US NLM:Pubmed: trial classification
bdanalytics
Date: (Tue) Apr 28, 2015
Introduction:
Data: Source: Training: https://courses.edx.org/c4x/MITx/15.071x_2/asset/clinical_trial.csv
New:
Time period:
Synopsis:
Based on analysis utilizing <> techniques,
[](.png)
Potential next steps include:
- Organization:
- Categorize by chunk
- Priority criteria:
- Ease of change
- Impacts report
- Cleans innards
- Bug report
- manage.missing.data chunk:
- cleaner way to manage re-splitting of training vs. new entity
- extract.features chunk:
- Convert user-specified mutate code to config specs
- fit.models chunk:
- Prediction accuracy scatter graph:
- Add tiles (raw vs. PCA)
- Use shiny for drop-down of “important” features
Use plot.ly for interactive plots ?
- Change .fit suffix of model metrics to .mdl if it’s data independent (e.g. AIC, Adj.R.Squared - is it truly data independent ?, etc.)
- move model_type parameter to myfit_mdl before indep_vars_vctr (keep all model_* together)
- create a custom model for rpart that has minbucket as a tuning parameter
varImp for randomForest crashes in caret version:6.0.41 -> submit bug report
- Probability handling for multinomials vs. desired binomial outcome
- ROCR currently supports only evaluation of binary classification tasks (version 1.0.7)
extensions toward multiclass classification are scheduled for the next release
- Skip trControl.method=“cv” for dummy classifier ?
- Add custom model to caret for a dummy (baseline) classifier (binomial & multinomial) that generates proba/outcomes which mimics the freq distribution of glb_rsp_var values; Right now glb_dmy_glm_mdl always generates most frequent outcome in training data
glm_dmy_mdl should use the same method as glm_sel_mdl until custom dummy classifer is implemented
- Compare glb_sel_mdl vs. glb_fin_mdl:
- varImp
- Prediction differences (shd be minimal ?)
- Move glb_analytics_diag_plots to mydsutils.R: (+) Easier to debug (-) Too many glb vars used
- Add print(ggplot.petrinet(glb_analytics_pn) + coord_flip()) at the end of every major chunk
- Parameterize glb_analytics_pn
- Move glb_impute_missing_data to mydsutils.R: (-) Too many glb vars used; glb_<>_df reassigned
- Replicate myfit_mdl_classification features in myfit_mdl_regression
- Do non-glm methods handle interaction terms ?
- f-score computation for classifiers should be summation across outcomes (not just the desired one ?)
- Add accuracy computation to glb_dmy_mdl in predict.data.new chunk
- Why does splitting fit.data.training.all chunk into separate chunks add an overhead of ~30 secs ? It’s not rbind b/c other chunks have lower elapsed time. Is it the number of plots ?
- Incorporate code chunks in print_sessionInfo
- Test against
- projects in github.com/bdanalytics
- lectures in jhu-datascience track
Analysis:
rm(list=ls())
set.seed(12345)
options(stringsAsFactors=FALSE)
source("~/Dropbox/datascience/R/mydsutils.R")
source("~/Dropbox/datascience/R/myplot.R")
source("~/Dropbox/datascience/R/mypetrinet.R")
# Gather all package requirements here
#suppressPackageStartupMessages(require())
#packageVersion("snow")
#require(sos); findFn("pinv", maxPages=2, sortby="MaxScore")
# Analysis control global variables
glb_trnng_url <- "https://courses.edx.org/c4x/MITx/15.071x_2/asset/clinical_trial.csv"
glb_newdt_url <- "<newdt_url>"
glb_is_separate_newent_dataset <- FALSE # or TRUE
glb_split_entity_newent_datasets <- TRUE # or FALSE
glb_split_newdata_method <- "sample" # "condition" or "sample" or "copy"
glb_split_newdata_condition <- "<col_name> <condition_operator> <value>" # or NULL
glb_split_newdata_size_ratio <- 0.3 # > 0 & < 1
glb_split_sample.seed <- 144 # or any integer
glb_max_trnent_obs <- NULL # or any integer
glb_drop_vars <- c(NULL) # or c("<col_name>")
glb_is_regression <- FALSE; glb_is_classification <- TRUE; glb_is_binomial <- TRUE
glb_rsp_var_raw <- "trial"
# for classification, the response variable has to be a factor
glb_rsp_var <- "trial.fctr"
# if the response factor is based on numbers e.g (0/1 vs. "A"/"B"),
# caret predict(..., type="prob") crashes
glb_map_rsp_raw_to_var <- function(raw) {
relevel(factor(ifelse(raw == 1, "Y", "N")), as.factor(c("Y", "N")), ref="N")
#as.factor(paste0("B", raw))
#as.factor(raw)
}
glb_map_rsp_raw_to_var(c(1, 1, 0, 0, 0))## [1] Y Y N N N
## Levels: N Yglb_map_rsp_var_to_raw <- function(var) {
as.numeric(var) - 1
#as.numeric(var)
#levels(var)[as.numeric(var)]
#c(" <=50K", " >50K")[as.numeric(var)]
}
glb_map_rsp_var_to_raw(glb_map_rsp_raw_to_var(c(1, 1, 0, 0, 0)))## [1] 1 1 0 0 0if ((glb_rsp_var != glb_rsp_var_raw) & is.null(glb_map_rsp_raw_to_var))
stop("glb_map_rsp_raw_to_var function expected")
glb_rsp_var_out <- paste0(glb_rsp_var, ".predict.") # model_id is appended later
glb_id_vars <- c(NULL) # or c("<id_var>")
glb_is_textual <- TRUE # or TRUE # vs. glb_is_numerical ???
#Sys.setlocale("LC_ALL", "C") # For english
glb_txt_vars <- c("title", "abstract") # or c("<col_name>")
glb_append_stop_words <- NULL # or c("<freq_word>")
glb_sprs_threshold <- 0.950 # Ideally, numrows(glb_feats_df) << numrows(glb_trnent_df)
# List transformed vars
glb_exclude_vars_as_features <- c(NULL) # or c("<var_name>")
if (glb_is_textual)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
glb_txt_vars)
if (glb_rsp_var_raw != glb_rsp_var)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
glb_rsp_var_raw)
# List feats that shd be excluded due to known causation by prediction variable
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c(NULL)) # or c("<col_name>")
# List output vars (useful during testing in console)
# glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
# grep(glb_rsp_var_out, names(glb_trnent_df), value=TRUE))
glb_impute_na_data <- FALSE # or TRUE
glb_mice_complete.seed <- 144 # or any integer
# rpart: .rnorm messes with the models badly
# caret creates dummy vars for factor feats which messes up the tuning
# - better to feed as.numeric(<feat>.fctr) to caret
# Regression
if (glb_is_regression)
glb_models_method_vctr <- c("lm", "glm", "rpart", "rf") else
# Classification
if (glb_is_binomial)
glb_models_method_vctr <- c("glm", "rpart", "rf") else
glb_models_method_vctr <- c("rpart", "rf")
# Baseline prediction model feature(s)
glb_Baseline_mdl_var <- NULL # or c("<col_name>")
glb_model_metric_terms <- NULL # or matrix(c(
# 0,1,2,3,4,
# 2,0,1,2,3,
# 4,2,0,1,2,
# 6,4,2,0,1,
# 8,6,4,2,0
# ), byrow=TRUE, nrow=5)
glb_model_metric <- NULL # or "<metric_name>"
glb_model_metric_maximize <- NULL # or FALSE (TRUE is not the default for both classification & regression)
glb_model_metric_smmry <- NULL # or function(data, lev=NULL, model=NULL) {
# confusion_mtrx <- t(as.matrix(confusionMatrix(data$pred, data$obs)))
# #print(confusion_mtrx)
# #print(confusion_mtrx * glb_model_metric_terms)
# metric <- sum(confusion_mtrx * glb_model_metric_terms) / nrow(data)
# names(metric) <- glb_model_metric
# return(metric)
# }
glb_tune_models_df <-
rbind(
#data.frame(parameter="cp", min=0.00005, max=0.00005, by=0.000005),
#seq(from=0.01, to=0.01, by=0.01)
#data.frame(parameter="mtry", min=2, max=4, by=1),
data.frame(parameter="dummy", min=2, max=4, by=1)
)
# or NULL
glb_n_cv_folds <- 3 # or NULL
glb_clf_proba_threshold <- NULL # 0.5
# Model selection criteria
if (glb_is_regression)
glb_model_evl_criteria <- c("min.RMSE.OOB", "max.R.sq.OOB", "max.Adj.R.sq.fit")
if (glb_is_classification) {
if (glb_is_binomial)
glb_model_evl_criteria <- c("max.Accuracy.OOB", "max.Kappa.OOB", "min.aic.fit") else
glb_model_evl_criteria <- c("max.Accuracy.OOB", "max.Kappa.OOB")
}
glb_sel_mdl_id <- NULL # or "<model_id_prefix>.<model_method>"
glb_fin_mdl_id <- glb_sel_mdl_id # or "Final"
glb_out_pfx <- "Pubmed_Trials_"
# Depict process
glb_analytics_pn <- petrinet(name="glb_analytics_pn",
trans_df=data.frame(id=1:6,
name=c("data.training.all","data.new",
"model.selected","model.final",
"data.training.all.prediction","data.new.prediction"),
x=c( -5,-5,-15,-25,-25,-35),
y=c( -5, 5, 0, 0, -5, 5)
),
places_df=data.frame(id=1:4,
name=c("bgn","fit.data.training.all","predict.data.new","end"),
x=c( -0, -20, -30, -40),
y=c( 0, 0, 0, 0),
M0=c( 3, 0, 0, 0)
),
arcs_df=data.frame(
begin=c("bgn","bgn","bgn",
"data.training.all","model.selected","fit.data.training.all",
"fit.data.training.all","model.final",
"data.new","predict.data.new",
"data.training.all.prediction","data.new.prediction"),
end =c("data.training.all","data.new","model.selected",
"fit.data.training.all","fit.data.training.all","model.final",
"data.training.all.prediction","predict.data.new",
"predict.data.new","data.new.prediction",
"end","end")
))
#print(ggplot.petrinet(glb_analytics_pn))
print(ggplot.petrinet(glb_analytics_pn) + coord_flip())## Loading required package: grid
glb_analytics_avl_objs <- NULL
glb_script_tm <- proc.time()
glb_script_df <- data.frame(chunk_label="import_data",
chunk_step_major=1, chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"])
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed import_data 1 0 0.002Step 1: import data
glb_entity_df <- myimport_data(url=glb_trnng_url,
comment=ifelse(!glb_is_separate_newent_dataset, "glb_entity_df", "glb_trnent_df"),
force_header=TRUE)## [1] "Reading file ./data/clinical_trial.csv..."
## [1] "dimensions of data in ./data/clinical_trial.csv: 1,860 rows x 3 cols"
## title
## 1 Treatment of Hodgkin's disease and other cancers with 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU; NSC-409962).
## 2 Cell mediated immune status in malignancy--pretherapy and post-therapy assessment.
## 3 Neoadjuvant vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophosphamide in early nonresponsive breast cancer: phase III randomized GeparTrio trial.
## 4 Randomized phase 3 trial of fluorouracil, epirubicin, and cyclophosphamide alone or followed by Paclitaxel for early breast cancer.
## 5 Is first-line single-agent mitoxantrone in the treatment of high-risk metastatic breast cancer patients as effective as combination chemotherapy? No difference in survival but higher quality of life were found in a multicenter randomized trial.
## 6 Expression of Bcl-2 in node-negative breast cancer is associated with various prognostic factors, but does not predict response to one course of perioperative chemotherapy.
## abstract
## 1
## 2 Twenty-eight cases of malignancies of different kinds were studied to assess T-cell activity and population before and after institution of therapy. Fifteen cases were diagnosed as non-metastasising squamous cell carcinoma of larynx, pharynx, laryngopharynx, hypopharynx and tonsils. Seven cases were non-metastasising infiltrating duct carcinoma of breast and 6 cases were non-Hodgkin's lymphoma (NHL). It was observed that 3 out of 15 cases (20%) of squamous cell carcinoma cases were Mantoux test (MT) negative with a T-cell population of less than 40%, 2 out of 7 cases (28.6%) of infiltrating duct carcinoma of breast were MT negative with a T-cell population of less than 40% and 3 out of 6 cases (50%) of NHL were MT negative with a T-cell population of less than 40%. The normal controls, consisting of apparently normal healthy adults, had a T-cell population of more than 40% and were all MT positive. The patients who showed a negative skin test and a T-cell population less than 40% were further subjected to assessment of T-cell population and activity after appropriate therapy, and clinical cure of the disease. It was observed that 2 out of 3 cases (66.66%) of squamous cell carcinomas, 2 out of 2 cases (100%) of adenocarcinomas and one out of 3 cases (33.33%) of NHL showed positive conversion with a T-cell population of more than 40%.
## 3 BACKGROUND: Among breast cancer patients, nonresponse to initial neoadjuvant chemotherapy is associated with unfavorable outcome. We compared the response of nonresponding patients who continued the same treatment with that of patients who switched to a well-tolerated non-cross-resistant regimen. METHODS: Previously untreated breast cancer patients received two 3-week cycles of docetaxel at 75 mg/m(2), doxorubicin at 50 mg/m(2), and cyclophosphamide at 500 mg/m(2) per day (TAC). Patients whose tumors did not decrease in size by at least 50% were randomly assigned to four additional cycles of TAC or to four cycles of vinorelbine at 25 mg/m(2) and capecitabine at 2000 mg/m(2) (NX). The outcome was sonographic response, defined as a reduction in the product of the two largest perpendicular diameters by at least 50%. A difference of 10% or less in the sonographic response qualified as noninferiority of the NX treatment. Pathological complete response was defined as no invasive or in situ residual tumor masses in the breast and lymph nodes. Toxic effects were assessed. All statistical tests were two-sided. RESULTS: Of 2090 patients enrolled in the GeparTrio study, 622 (29.8%) who did not respond to two initial cycles of TAC were randomly assigned to an additional four cycles of TAC (n = 321) or to four cycles of NX (n = 301). Sonographic response rate was 50.5% for the TAC arm and 51.2% for the NX arm. The difference of 0.7% (95% confidence interval = -7.1% to 8.5%) demonstrated noninferiority of NX (P = .008). Similar numbers of patients in both arms received breast-conserving surgery (184 [57.3%] in the TAC arm vs 180 [59.8%] in the NX arm) and had a pathological complete response (5.3% vs 6.0%). Fewer patients in the NX arm than in the TAC arm had hematologic toxic effects, mucositis, infections, and nail changes, but more had hand-foot syndrome and sensory neuropathy. CONCLUSION: Pathological complete responses to both regimens were marginal. Among patients who did not respond to the initial neoadjuvant TAC treatment, similar efficacy but better tolerability was observed by switching to NX than continuing with TAC.
## 4 BACKGROUND: Taxanes are among the most active drugs for the treatment of metastatic breast cancer, and, as a consequence, they have also been studied in the adjuvant setting.METHODS: After breast cancer surgery, women with lymph node-positive disease were randomly assigned to treatment with fluorouracil, epirubicin, and cyclophosphamide (FEC) or with FEC followed by weekly paclitaxel (FEC-P). The primary endpoint of study-5-year disease-free survival (DFS)-was assessed by Kaplan-Meier analysis. Secondary endpoints included overall survival and analysis of the prognostic and predictive value of clinical and molecular (hormone receptors by immunohistochemistry and HER2 by fluorescence in situ hybridization) markers. Associations and interactions were assessed with a multivariable Cox proportional hazards model for DFS for the following covariates: age, menopausal status, tumor size, lymph node status, type of chemotherapy, tumor size, positive lymph nodes, HER2 status, and hormone receptor status. All statistical tests were two-sided.RESULTS: Among the 1246 eligible patients, estimated rates of DFS at 5 years were 78.5% in the FEC-P arm and 72.1% in the FEC arm (difference = 6.4%, 95% confidence interval [CI] = 1.6% to 11.2%; P = .006). FEC-P treatment was associated with a 23% reduction in the risk of relapse compared with FEC treatment (146 relapses in the 614 patients in the FEC-P arm vs 193 relapses in the 632 patients in the FEC arm, hazard ratio [HR] = 0.77, 95% CI = 0.62 to 0.95; P = .022) and a 22% reduction in the risk of death (73 and 95 deaths, respectively, HR = 0.78, 95% CI = 0.57 to 1.06; P = .110). Among the 928 patients for whom tumor samples were centrally analyzed, type of chemotherapy (FEC vs FEC-P) (P = .017), number of involved axillary lymph nodes (P < .001), tumor size (P = .020), hormone receptor status (P = .004), and HER2 status (P = .006) were all associated with DFS. We found no statistically significant interaction between HER2 status and paclitaxel treatment or between hormone receptor status and paclitaxel treatment.CONCLUSIONS: Among patients with operable breast cancer, FEC-P treatment statistically significantly reduced the risk of relapse compared with FEC as adjuvant therapy.
## 5 BACKGROUND: To determine whether patients with high-risk metastatic breast cancer draw benefit from combination chemotherapy as first-line treatment. PATIENTS AND METHODS: A total of 260 women with measurable metastatic breast cancer fulfilling high-risk criteria, previously untreated with chemotherapy for their metastatic disease, were randomized to receive either mitoxantrone 12 mg/m(2) or the combination of fluorouracil 500 mg/m(2), epirubicin 50 mg/m(2) and cyclophosphamide 500 mg/m(2) (FEC) every 3 weeks. Treatment was continued until complete remission plus two cycles, or until disease progression. In the case of partial remission or stable disease, treatment was stopped after 12 cycles. Second-line treatment was vindesine, mitomycin and prednisolone. Gain from treatment was estimated using a modified Brunner's score composed of time to progression, patients' rating of the treatment benefit, alopecia, vomiting and performance status. RESULTS: After recruitment from 1992 to 1997 and observation from 1997 to 1999, the final evaluation showed that single-agent treatment with mitoxantrone does not differ significantly from combination treatment with FEC in terms of response, objective remission rate, remission duration, time to response, time to best response, time to progression or overall survival. There was, however, a significant difference in gain from treatment using a modified Brunner's score favoring the single-agent treatment arm. There was no evidence that any subgroup would fare better with combination treatment. CONCLUSIONS: No significant difference was detected between the treatment with mitoxantrone as a single agent and the combination of low-dose FEC in terms of response or survival; therefore, the imperative of the necessity of first-line combination chemotherapy for patients with high-risk metastatic breast cancer may be questioned. Since toxicity and quality of life score favored the single-agent mitoxantrone treatment arm, this treatment may be offered to patients preferring quality of life to a potential small prolongation of survival.
## 6 The aim of this study was to assess relationships between Bcl-2 expression, response to chemotherapy and a number of pathological and biological tumour parameters in premenopausal, lymph node-negative breast cancer patients. Expression of Bcl-2 was determined using immunohistochemistry on paraffin-embedded sections in a series of 441 premenopausal, lymph node-negative breast cancers of patients randomised to receive perioperative chemotherapy (5-fluorouracil, doxorubicin, cyclophosphamide) or no perioperative chemotherapy. Immunohistochemistry of Bcl-2 was evaluated by scoring both staining intensity (0-3) and number of positive cells (0-2). Using these scores tumours were grouped into categories 0-6. It was found that 9.2% of the tumours were completely negative (0), 17.2% weakly (1 + 2), 41.6% moderately (3 + 4) and 31.9% strongly positive (5 + 6) for Bcl-2. A positive correlation was found between high Bcl-2 expression and oestrogen (P < 0.001) and progesterone receptor positivity (P < 0.001) and low tumour grade (P < 0.001), whereas high Bcl-2 expression was negatively correlated with p53 (P < 0.001) and c-erb-B-2 positively (P < 0.001), high Ki-67 index (P < 0.001), mitotic index (P < 0.001) and large tumour size (P = 0.006). Patients with tumours expressing high levels of Bcl-2 (overall score 3-6) had a significantly better disease-free (P = 0.004) and overall (P = 0.009) survival. However, in a multivariate model this association no longer remained significant. There was a trend for an effect of adjuvant chemotherapy on disease-free survival both for patients with Bcl-2-positive (HR-0.61, 95% CI 0.35-1.06, P = 0.07) and negative (HR = 0.55, 95% CI 0.27-1.12, P = 0.09) breast tumours at a median follow-up of 49 months. The level of Bcl-2 expression does not seem to predict response to perioperative chemotherapy in premenopausal, lymph node-negative breast cancer patients. High levels of Bcl-2 are preferentially expressed in well-differentiated tumours and are associated with favourable prognosis. However, Bcl-2 expression is not an independent prognostic factor in this patient series.
## trial
## 1 1
## 2 0
## 3 1
## 4 1
## 5 1
## 6 0
## title
## 65 A phase I trial of topical topitriol (calcitriol, 1,25-dihydroxyvitamin D3) to prevent chemotherapy-induced alopecia.
## 283 Population pharmacokinetics and pharmacodynamics of oral etoposide.
## 605 Positron emission tomography studies in patients with locally advanced and/or metastatic breast cancer: a method for early therapy evaluation?
## 947 Xeloda and Taxotere: a review of the development of the combination for use in metastatic breast cancer.
## 1353 Efficacy of pamidronate in reducing skeletal complications in patients with breast cancer and lytic bone metastases. Protocol 19 Aredia Breast Cancer Study Group.
## 1838 Phase II trial of carboplatin in advanced breast carcinoma: a Cancer and Leukemia Group B Study.
## abstract
## 65 This study evaluated the toxicity and efficacy of topical topitriol (calcitriol, 1,25-dihydroxyvitamin D3) to prevent chemotherapy-induced alopecia (CIA). Patients with breast cancer scheduled to receive FAC chemotherapy (5-fluorouracil, adriamycin and cyclophosphamide) were eligible for the study. Initially, the first six patients were randomized in a double-blind fashion to have received topitriol or placebo with all subsequent patients being treated with topitriol. Topitriol cream (0.0025 or 0.005%; 25 and 50 microg/g concentration) was administered topically twice a day. Three different doses and schedules of administration were evaluated including: 500 and 1000 microg daily for 7 days prior to chemotherapy, and 2000 microg daily for 5 days prior and 5 days post-chemotherapy. Fourteen patients were treated (12 with topitriol and two with placebo) at three different dose levels. All patients developed grade 2 alopecia between day 20 and 30 after chemotherapy, demonstrating the lack of efficacy of topical topitriol on this schedule of administration to prevent CIA. Eight patients exposed to topitriol developed a toxic maculopapular dermatitis in areas exposed to the drug. In conclusion, topical topitriol at the doses and schedules evaluated in this trial was ineffective to prevent CIA and induced a local dermatitis in areas exposed to the drug.
## 283 AIMS: To study the population pharmacokinetics and pharmacodynamics of oral etoposide in patients with solid tumours. METHODS: A prospective, open label, cross-over, bioavailability study was performed in 50 adult patients with miscellaneous, advanced stage solid tumours, who were receiving oral (100 mg capsules) etoposide for 14 days and i.v. (50 mg) etoposide on day 1 or day 7 in randomised order during the first cycle treatment. Total and unbound etoposide concentration were assayed by h.p.l.c. Population PK parameters estimation was done by using the P-Pharm software (Simed). Haematological toxicity and tumour response were the main pharmacodynamic endpoints. RESULTS: Mean clearance was 1.14 l h(-1) (CV 25%). Creatinine clearance was the only covariable to significantly reduce clearance variability (residual CV 18%). (CL = 0.74 + 0.0057 CLCR; r(2) = 0.32). Mean bioavailability was 45% (CV 22%) and mean protein binding 91.5% (CV 5%). Exposure to free, pharmacologically active etoposide (free AUC p.o.) was highly variable (mean value 2.8 mg l(-1) h; CV 64%; range 0.4-9.5). It decreased with increased creatinine clearance and increased with age which accounted for 9% of the CV. Mean free AUC p.o. was the best predictor of neutropenia. Free AUC50 (exposure producing a 50% reduction in absolute neutrophil count) was 1.80 mg l(-1) h. In patients with lung cancer, the free AUC p.o. was higher in the two patients with responsive tumour (5.9 mg l(-1) h) than in patients with stable (2.1 mg l-1 h) or progressive disease (2.3 mg l-1 h) (P = 0.01). CONCLUSIONS: Exposure to free etoposide during prolonged oral treatment is highly variable and is the main determinant of pharmacodynamic effects. The population PK model based on creatinine clearance is poorly predictive of exposure. Therapeutic drug monitoring would be necessary for dose individualization or to study the relationship between exposure and antitumour effect.
## 605 PURPOSE: To investigate if sequential positron emission tomographic (PET) scans with the glucose analog 18F-2-fluoro-2-deoxy-D-glucose (18FDG) and/or L-methyl-11C-methionine (11C-methionine) in patients with breast cancer could provide early information on the efficacy of polychemotherapy. PATIENTS AND METHODS: Sixteen patients with breast cancer (11 with locally advanced tumors, three with recurrent disease in the contralateral breast, two of them with distant and regional metastases, and two with distant metastases) underwent a baseline and two follow-up PET scans after the first and third/fourth polychemotherapy course. Tumor response was determined clinically/radiographically after three/four polychemotherapy courses. RESULTS: Five patients were investigated with 18FDG, seven with both 11C-methionine and 18FDG, and four with only 11C-methionine before polychemotherapy. 11C-methionine presented a more distinct visualization of primary/contralateral breast cancers in five of seven patients when compared with 18FDG. Twelve of 16 patients demonstrated a response using conventional methods after the third/fourth course of polychemotherapy. Eight of these 12 clinical responders had a significant decrease in tracer uptake at the first PET scan performed 6 to 13 days after the first polychemotherapy course, and these reductions were further augmented after the third/fourth course and corresponded to the conventional therapy evaluation (clinical examination, computed tomography [CT], ultrasonography, and mammography). CONCLUSION: Our data indicate that PET may be of clinical value in predicting response to chemotherapy in patients with locally advanced breast cancer and/or metastatic disease earlier than any other method used.
## 947 Docetaxel (Taxotere) and capecitabine (Xeloda) demonstrate synergy in pre-clinical studies and both show high single-agent efficacy in metastatic breast cancer patients previously treated with an anthracycline. The two drugs used in combination in a similar group of patients have been evaluated in a randomised phase III trial comparing this combination with single-agent docetaxel. The results of this study indicate the combination arm to be favourable with significantly superior median time to disease progression (6.1 vs 4.2 months, hazard ratio 0.652, p=0.0001), objective response rate and overall survival without a significant increase in the overall incidence of treatment related adverse effects. This combination of docetaxel and capecitabine provides improved survival prospects for patients with metastatic breast cancer. Ongoing studies will provide further data to optimise the scheduling and use of this combination in the future.
## 1353 BACKGROUND: Bisphosphonates such as pamidronate disodium inhibit osteoclast-induced bone resorption associated with cancer that has metastasized to bone. METHODS: Women with stage IV breast cancer who were receiving cytotoxic chemotherapy and had at least one lytic bone lesion were given either placebo or pamidronate (90 mg) as a two-hour intravenous infusion monthly for 12 cycles. Skeletal complications, including pathologic fractures, the need for radiation to bone or bone surgery, spinal cord compression, and hypercalcemia (a serum calcium concentration above 12 mg per deciliter [3.0 mmol per liter] or elevated to any degree and requiring treatment), were assessed monthly. Bone pain, use of analgesic drugs, performance status, and quality of life were assessed throughout the trial. RESULTS: The efficacy of treatment was evaluated in 380 of 382 randomized patients, 185 receiving pamidronate and 195 receiving placebo. The median time to the occurrence of the first skeletal complication was greater in the pamidronate group than in the placebo group (13.1 vs. 7.0 months, P=0.005), and the proportion of patients in whom any skeletal complication occurred was lower (43 percent vs. 56 percent, P = 0.008). There was significantly less increase in bone pain (P=0.046) and deterioration of performance status (P=0.027) in the pamidronate group than in the placebo group. Pamidronate was well tolerated. CONCLUSIONS: Monthly infusions of pamidronate as a supplement to chemotherapy can protect against skeletal complications in women with stage IV breast cancer who have osteolytic bone metastases.
## 1838
## trial
## 65 0
## 283 1
## 605 0
## 947 0
## 1353 0
## 1838 1
## title
## 1855 Intra-arterial cytotoxic therapy and radiotherapy for locally advanced breast cancer. Interim report of a trial.
## 1856 Clinical trial of 2-Br- -ergocryptine (CB154) in advanced breast cancer.
## 1857 Five-drug therapy for advanced breast cancer: a phase I study.
## 1858 A controlled clinical trial to compare transethidal hypophysectomy with yttrium implant of the pituitary in the treatment of advanced breast cancer.
## 1859 A clinical trial to compare early and late pituitary ablation in advanced cancer of the breast.
## 1860 A clinical trial of tri-iodothyronine as a hormone potentiator in advanced breast cancer.
## abstract trial
## 1855 0
## 1856 0
## 1857 1
## 1858 0
## 1859 0
## 1860 0
## 'data.frame': 1860 obs. of 3 variables:
## $ title : chr "Treatment of Hodgkin's disease and other cancers with 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU; NSC-409962)." "Cell mediated immune status in malignancy--pretherapy and post-therapy assessment." "Neoadjuvant vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophosphamide in early nonresponsive breast cancer: phase "| __truncated__ "Randomized phase 3 trial of fluorouracil, epirubicin, and cyclophosphamide alone or followed by Paclitaxel for early breast can"| __truncated__ ...
## $ abstract: chr "" "Twenty-eight cases of malignancies of different kinds were studied to assess T-cell activity and population before and after in"| __truncated__ "BACKGROUND: Among breast cancer patients, nonresponse to initial neoadjuvant chemotherapy is associated with unfavorable outcom"| __truncated__ "BACKGROUND: Taxanes are among the most active drugs for the treatment of metastatic breast cancer, and, as a consequence, they "| __truncated__ ...
## $ trial : int 1 0 1 1 1 0 1 0 0 0 ...
## - attr(*, "comment")= chr "glb_entity_df"
## NULLif (!glb_is_separate_newent_dataset) {
glb_trnent_df <- glb_entity_df; comment(glb_trnent_df) <- "glb_trnent_df"
} # else glb_entity_df is maintained as is for chunk:inspectORexplore.data
if (glb_is_separate_newent_dataset) {
glb_newent_df <- myimport_data(
url=glb_newdt_url,
comment="glb_newent_df", force_header=TRUE)
# To make plots / stats / checks easier in chunk:inspectORexplore.data
glb_entity_df <- rbind(glb_trnent_df, glb_newent_df); comment(glb_entity_df) <- "glb_entity_df"
} else {
if (!glb_split_entity_newent_datasets) {
stop("Not implemented yet")
glb_newent_df <- glb_trnent_df[sample(1:nrow(glb_trnent_df),
max(2, nrow(glb_trnent_df) / 1000)),]
} else if (glb_split_newdata_method == "condition") {
glb_newent_df <- do.call("subset",
list(glb_trnent_df, parse(text=glb_split_newdata_condition)))
glb_trnent_df <- do.call("subset",
list(glb_trnent_df, parse(text=paste0("!(",
glb_split_newdata_condition,
")"))))
} else if (glb_split_newdata_method == "sample") {
require(caTools)
set.seed(glb_split_sample.seed)
split <- sample.split(glb_trnent_df[, glb_rsp_var_raw],
SplitRatio=(1-glb_split_newdata_size_ratio))
glb_newent_df <- glb_trnent_df[!split, ]
glb_trnent_df <- glb_trnent_df[split ,]
} else if (glb_split_newdata_method == "copy") {
glb_trnent_df <- glb_entity_df
comment(glb_trnent_df) <- "glb_trnent_df"
glb_newent_df <- glb_entity_df
comment(glb_newent_df) <- "glb_newent_df"
} else stop("glb_split_newdata_method should be %in% c('condition', 'sample', 'copy')")
comment(glb_newent_df) <- "glb_newent_df"
myprint_df(glb_newent_df)
str(glb_newent_df)
if (glb_split_entity_newent_datasets) {
myprint_df(glb_trnent_df)
str(glb_trnent_df)
}
} ## Loading required package: caTools## title
## 4 Randomized phase 3 trial of fluorouracil, epirubicin, and cyclophosphamide alone or followed by Paclitaxel for early breast cancer.
## 8 Long-term endometrial effects in postmenopausal women with early breast cancer participating in the Intergroup Exemestane Study (IES)--a randomised controlled trial of exemestane versus continued tamoxifen after 2-3 years tamoxifen.
## 9 Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82c randomised trial.
## 11 Adjuvant therapy of cancer via the cellular immune mechanism or fibrin by induced fibrinolysis and oral anticoagulants.
## 19 Results of two randomized trials evaluating adjuvant anthracycline-based chemotherapy in 1146 patients with early breast cancer.
## 31 Pathological complete response rates comparing 3 versus 6 cycles of epidoxorubicin and docetaxel in the neoadjuvant setting of patients with stage II and III breast cancer.
## abstract
## 4 BACKGROUND: Taxanes are among the most active drugs for the treatment of metastatic breast cancer, and, as a consequence, they have also been studied in the adjuvant setting.METHODS: After breast cancer surgery, women with lymph node-positive disease were randomly assigned to treatment with fluorouracil, epirubicin, and cyclophosphamide (FEC) or with FEC followed by weekly paclitaxel (FEC-P). The primary endpoint of study-5-year disease-free survival (DFS)-was assessed by Kaplan-Meier analysis. Secondary endpoints included overall survival and analysis of the prognostic and predictive value of clinical and molecular (hormone receptors by immunohistochemistry and HER2 by fluorescence in situ hybridization) markers. Associations and interactions were assessed with a multivariable Cox proportional hazards model for DFS for the following covariates: age, menopausal status, tumor size, lymph node status, type of chemotherapy, tumor size, positive lymph nodes, HER2 status, and hormone receptor status. All statistical tests were two-sided.RESULTS: Among the 1246 eligible patients, estimated rates of DFS at 5 years were 78.5% in the FEC-P arm and 72.1% in the FEC arm (difference = 6.4%, 95% confidence interval [CI] = 1.6% to 11.2%; P = .006). FEC-P treatment was associated with a 23% reduction in the risk of relapse compared with FEC treatment (146 relapses in the 614 patients in the FEC-P arm vs 193 relapses in the 632 patients in the FEC arm, hazard ratio [HR] = 0.77, 95% CI = 0.62 to 0.95; P = .022) and a 22% reduction in the risk of death (73 and 95 deaths, respectively, HR = 0.78, 95% CI = 0.57 to 1.06; P = .110). Among the 928 patients for whom tumor samples were centrally analyzed, type of chemotherapy (FEC vs FEC-P) (P = .017), number of involved axillary lymph nodes (P < .001), tumor size (P = .020), hormone receptor status (P = .004), and HER2 status (P = .006) were all associated with DFS. We found no statistically significant interaction between HER2 status and paclitaxel treatment or between hormone receptor status and paclitaxel treatment.CONCLUSIONS: Among patients with operable breast cancer, FEC-P treatment statistically significantly reduced the risk of relapse compared with FEC as adjuvant therapy.
## 8 BACKGROUND: The antiestrogen tamoxifen may have partial estrogen-like effects on the postmenopausal uterus. Aromatase inhibitors (AIs) are increasingly used after initial tamoxifen in the adjuvant treatment of postmenopausal early breast cancer due to their mechanism of action: a potential benefit being a reduction of uterine abnormalities caused by tamoxifen.PATIENTS AND METHODS: Sonographic uterine effects of the steroidal AI exemestane were studied in 219 women participating in the Intergroup Exemestane Study: a large trial in postmenopausal women with estrogen receptor-positive (or unknown) early breast cancer, disease free after 2-3 years of tamoxifen, randomly assigned to continue tamoxifen or switch to exemestane to complete 5 years adjuvant treatment. The primary end point was the proportion of patients with abnormal (> or =5 mm) endometrial thickness (ET) on transvaginal ultrasound 24 months after randomisation.RESULTS: The analysis included 183 patients. Two years after randomisation, the proportion of patients with abnormal ET was significantly lower in the exemestane compared with tamoxifen arm (36% versus 62%, respectively; P = 0.004). This difference emerged within 6 months of switching treatment (43.5% versus 65.2%, respectively; P = 0.01) and disappeared within 12 months of treatment completion (30.8% versus 34.7%, respectively; P = 0.67).CONCLUSION: Switching from tamoxifen to exemestane significantly reverses endometrial thickening associated with continued tamoxifen.
## 9 BACKGROUND: Postmastectomy radiotherapy is associated with a lower locoregional recurrence rate and improved disease-free and overall survival when combined with chemotherapy in premenopausal high-risk breast-cancer patients. However, whether the same benefits apply also in postmenopausal women treated with adjuvant tamoxifen for similar high-risk cancer is unclear. In a randomised trial among postmenopausal women who had undergone mastectomy, we compared adjuvant tamoxifen alone with tamoxifen plus postoperative radiotherapy. METHODS: Between 1982 and 1990, postmenopausal women with high-risk breast cancer (stage II or III) were randomly assigned adjuvant tamoxifen (30 mg daily for 1 year) alone (689) or with postoperative radiotherapy to the chest wall and regional lymph nodes (686). Median follow-up was 123 months. The endpoints were first site of recurrence (locoregional recurrence, distant metastases, or both), and disease-free and overall survival. FINDINGS: Locoregional recurrence occurred in 52 (8%) of the radiotherapy plus tamoxifen group and 242 (35%) of the tamoxifen only group (p<0.001). In total there were 321 (47%) and 411 (60%) recurrences, respectively. Disease-free survival was 36% in the radiotherapy plus tamoxifen group and 24% in the tamoxifen alone group (p<0.001). Overall survival was also higher in the radiotherapy group (385 vs 434 deaths; survival 45 vs 36% at 10 years, p=0.03). INTERPRETATION: Postoperative radiotherapy decreased the risk of locoregional recurrence and was associated with improved survival in high-risk postmenopausal breast-cancer patients after mastectomy and limited axillary dissection, with 1 year of adjuvant tamoxifen treatment. Improved survival in high-risk breast cancer can best be achieved by a strategy of both locoregional and systemic tumour control.
## 11 The value of the oral anticoagulant warfarin sodium and fibrinolytic agents is discussed in relation to cancer surgery. A controlled trial of 128 patients showed that in a variety of recurrent cases the addition of warfarin to chemotherapy doubled the 2-year survival rate. The best results were obtained in postmenopausal patients with breast cancer. Warfarin depresses cellular immune responses which might militate against its use for cases undergoing "curative" surgery. Instead, induction of fibrinolysis by streptokinase or Brinase is suggested, because it increases the activity of the cellular immune mechanism. The results to date of an ongoing controlled randomized trial of streptokinase with surgery of tumors of the large bowel are presented, showing that the trends are in favor of streptokinase therapy; however, insufficient time has elapsed to make it, as yet, statistically significant. The action of streptokinase-induced plasmin and Brinase on lymphocytes is described.
## 19 Two randomized trials evaluated the effect of 6 courses of anthracycline-based chemotherapy in early breast cancer. A total of 1146 patients were included: 311 high-risk node-negative premenopausal patients and 835 high-risk node-negative or node-positive postmenopausal patients. Patients were randomized after surgery to receive either no chemotherapy (control group) or 6 courses of anthracycline-based chemotherapy (CT group). Postmenopausal patients received adjuvant tamoxifen for at least two years. Radiotherapy was delivered after completion of chemotherapy in the CT group. The 10-year disease-free survival (DFS) rates were 60% in the control group and 65% in the CT group (log-rank test, p = 0.01). The 10-year distant metastasis rates were 28% and 23% (p = 0.02), and the 10-year local recurrence rates were 12% and 10%, respectively (p = 0.24). Chemotherapy was significantly less effective in post-menopausal patients with estrogen receptor-positive tumors. Adjuvant anthracycline-based chemotherapy yielded a significant benefit for DFS by lowering the risk of distant metastases. After up to 10 years of follow-up, deferring radiotherapy after chemotherapy did not compromise local control.
## 31 We conducted a prospective randomized study to compare the results of 3 cycles of epidoxorubicin/docetaxel to 6 cycles of epidoxorubicin/docetaxel prior to surgery in breast cancer patients with clinical stages II and III. Forty-five patients eligible for neoadjuvant chemotherapy were randomly assigned to receive either 3 (group 1) or 6 (group 2) cycles of epidoxorubicin/docetaxel prior to surgery. Chemotherapy consisted of epidoxorubicin 75 mg/m and docetaxel 75 mg/m on day 1 in 3-week cycles. The primary endpoint was the pathological complete response (pCR) rate; secondary endpoints were the rates of breast-conserving surgery and the axillary lymph node status in both groups. A pCR occurred in 10% (two of 20) in Group 1 and in 36% (nine of 25) in Group 2, which was statistically significant (p=0.045). Breast-conserving surgery could be performed in 70% (14 of 20) in Group 1 and in 76% (19 of 25) in Group 2 (p=0.065). Axillary lymph node status was negative in 45% (nine of 20) in Group 1 and 52% (13 of 25) in Group 2 (p=0.86). We conclude that 6 cycles of pre-operative epidoxorubicin/docetaxel versus 3 cycles of pre-operative epidoxorubicin/docetaxel significantly increases the pCR rates for breast cancer patients.
## trial
## 4 1
## 8 0
## 9 0
## 11 0
## 19 1
## 31 1
## title
## 362 Effect of age and radiation dose on local control after breast conserving treatment: EORTC trial 22881-10882.
## 952 Adjuvant therapy with tamoxifen in operable breast cancer. 10 year results of the Naples (GUN) study.
## 1213 Repeated significance tests on accumulating survival data.
## 1624 Capecitabine combined with weekly docetaxel in Chinese patients > 65 years with anthracycline-resistant metastatic breast cancer.
## 1767 Trastuzumab plus capecitabine and docetaxel as first-line therapy for HER2-positive metastatic breast cancer: phase II results.
## 1847 Phase II study of mitolactol in chemotherapy-refractory metastatic breast cancer.
## abstract
## 362 PURPOSE: To determine whether the effect of an additional "boost" radiation after breast conservative therapy (BCT) on local control depends on age and evaluate the impact of a treatment policy with a threshold for age. PATIENTS AND METHODS: We used data from EORTC 22881-10882 trial, with median follow-up of 77.4 months. Patients receiving BCT and 50Gy whole breast irradiation were randomized to no boost and 16Gy boost (N=5318). RESULTS: In univariate analysis, a boost reduced local failure by a factor of 2 (P<0.0001). Multivariate analysis showed local control increased with age (P=0.0003). There was no evidence that the relative effect of a boost on local control depends on age (P=0.97) However in younger patients the 5-year local failure was higher, therefore the absolute reduction was greater. If the threshold-age for boost treatment were set at 40 years, 8.4% of the study population would receive a boost, resulting in a 5-year local failure of 6.1% in the study population. Changing the threshold-age to 60 years, 67% of the study population would receive a boost and the 5-year local failure would be reduced to 4.4%. CONCLUSIONS: In younger patients a boost dose resulted in a greater absolute reduction of local failure. The relative risk reduction was however similar for all ages. Applying a treatment policy with a threshold-age of 60 would result in 0.6% increase in local failure in the total study population, while sparing the boost to 1/3 of the patients.
## 952 Treatment with tamoxifen (TM), alone or in combination with cyclophosphamide, methotrexate, and fluorouracil (CMF), was used as an adjuvant to surgery in 433 patients with stage I, II, or III(T3a) breast cancer. Oestrogen receptors (ER) and progesterone (PgR) receptors were assayed in most cases. 308 premenopausal node-negative and postmenopausal node-negative or node-positive patients were randomised to receive TM, 30 mg daily for 2 years, or no further therapy. 125 premenopausal node-positive patients were randomised to receive either CMF for nine courses plus TM or CMF alone. After a median follow-up of 63 months TM significantly reduced the incidence of relapses and deaths compared with no therapy. A significant interaction between treatment effect and ER/PgR status was seen. Disease-free and overall survival were similar after treatment with CMF+ TM or CMF.
## 1213 The aim of the study was to compare the properties of two well-known group sequential methods and to demonstrate the effect of performing interim analyses on accumulating survival data without making appropriate adjustments of the nominal significance level. The properties of a group sequential method with fixed nominal significance level (Pocock stopping boundaries) and a method with increasing nominal level with each interim analysis (O'Brien-Fleming boundaries) were compared by stochastic simulation. Simulation experiments with survival times sampled from a breast cancer trial and from exponential distributions were performed. The true overall significance level with unplanned interim analyses increased from 5% to 14% when a maximum of five tests were performed. Both group sequential methods maintained the desired overall significance level. The O'Brien-Fleming method had higher power than Pocock's method. It also reduced the risk of early stopping based on immature data and should usually be preferred.
## 1624 BACKGROUND: There are no data on more tolerable capecitabine doses in elderly patients in Chinese population. The aim of this study was to evaluate the activity and safety of capecitabine combined with weekly docetaxel for the treatment of anthracycline-resistant metastatic breast cancer (MBC) in older Chinese patients.METHODS: MBC patients aged > 65 years pretreated with 1 - 5 prior chemotherapy regimens, including an anthracycline, received oral capecitabine 825 mg/m(2) twice daily, days 1 - 14, plus docetaxel 30 mg/m(2) on days 1 and 8 every 21 days. All 41 enrolled patients received at least 1 dose of treatment and were evaluable for safety; 38 received at least 2 cycles (median 4, range 2 - 8) and were evaluable for efficacy.RESULTS: The overall objective response rate was 47%, including complete responses in 8% of patients. Median time to progression was 8.9 months. Median overall survival was 17.6 months. The most common side effects were haematological and gastrointestinal toxicities and hand-foot syndrome. The only grade 3/4 adverse events were neutropenia (12%), alopecia (7%), grade 3 nausea and vomiting (2%) and grade 3 nail toxicity (2%).CONCLUSIONS: Capecitabine 825 mg/m(2) twice daily plus weekly docetaxel is active with an acceptable safety profile in Chinese women > 65 years with anthracycline-resistant MBC. Efficacy and tolerability compare favourably with previously reported trials evaluating higher capecitabine doses in combination with 3-weekly or weekly docetaxel.
## 1767 In human epidermal growth factor 2 (HER-2)-positive advanced breast cancer, taxanes plus trastuzumab are among the most widely applied options in the first-line setting. The addition of capecitabine to docetaxel significantly improves overall survival in anthracycline-pretreated metastatic breast cancer. We evaluated the efficacy and tolerability of trastuzumab plus capecitabine and docetaxel regimen as first-line therapy.PATIENTS AND METHODS: HER-2-positive patients who had received adjuvant anthracyclines received docetaxel at 75 mg/m(2) on day 1 and capecitabine 950 mg/m(2)/day, days 1-14, every 3 weeks until disease progression or unacceptable toxicity. Trastuzumab was administered at a dose of 6 mg/kg every 3 weeks. Time to progression (TTP) was defined as the primary end point. RESULTS: Twenty-nine patients were evaluable (median age 52, range 34-70 years). The regimen achieved objective responses in 11 patients (38%), including complete response in three (10.3%) and partial response in eight (27.5%). The median overall survival time was 25.5 months, and the median progression-free survival time was 7.8 months. The safety profile of the combination was favorable and predictable, with a low incidence of grade 3/4 adverse events. The most common adverse events were hand-foot syndrome, and gastrointestinal toxicities. Severe myelosuppression was rare and cardiac toxicity did not occur. CONCLUSION: These data confirm that the combination of trastuzumab plus capecitabine and docetaxel is highly active in patients with HER-2-overexpressing anthracycline-pretreated breast cancer, offering a significant survival benefit and is well tolerated.
## 1847 Thirty-eight evaluable patients with metastatic breast cancer refractory to hormonal therapy and multiple chemotherapy regimens were treated with mitolactol at a dose of 130 mg/m2/day orally for 10 days every 6 weeks. Only one patient, with nodal and chest wall metastases, had a sustained complete regression; two patients had stable disease; and 35 patients had disease progression. The toxicity, which was primarily hematologic, was acceptable.
## trial
## 362 0
## 952 1
## 1213 0
## 1624 1
## 1767 1
## 1847 1
## title
## 1842 Randomized trial of estrogen vs. tamoxifen therapy for advanced breast cancer.
## 1843 A clinical trial of aminoglutethimide in advanced postmenopausal breast carcinoma: low response in patients previously treated with medroxyprogesterone.
## 1847 Phase II study of mitolactol in chemotherapy-refractory metastatic breast cancer.
## 1852 Clinical trial of combined hormone therapy (MAP + TAM) for metastatic breast cancer.
## 1854 A comparative trial of transsphenoidal hypophysectomy and estrogen suppression with aminoglutethimide in advanced breast cancer.
## 1855 Intra-arterial cytotoxic therapy and radiotherapy for locally advanced breast cancer. Interim report of a trial.
## abstract
## 1842 Forty-three postmenopausal females with advanced breast cancer were studied in a prospective comparative trial of estrogen vs. an anti-estrogen (tamoxifen) therapy with a crossover to the alternative hormone with progressive disease. Ten of 19 patients (53%) responded to primary tamoxifen therapy and six of 24 (25%) responded to primary estrogen therapy. Crossover responses were observed in seven of 19 (37%) on the secondary tamoxifen therapy and in two of 10 (20%) on secondary estrogen therapy, and were not related to the response to the primary hormonal maneuver. Responses were related to the presence of estrogen receptor protein (ERP), particularly for tamoxifen therapy, although responses were observed in three of six ERP negative patients receiving estrogen and in seven of 25 (28%) of patients with an unknown ERP status. Complications were observed in 35 instances with estrogen therapy and in only five instances with tamoxifen therapy. Initial hormonal therapy with tamoxifen in postmenopausal patients with advanced breast cancer and ERP status positive or unknown is superior to primary estrogen treatment. Secondary therapy and response to estrogen or tamoxifen is not necessarily predicted by the initial hormone response, and crossover to the alternative therapy is generally indicated.
## 1843 In an attempt to define the influence of prior hormonal treatments upon aminoglutethimide activity in advanced cancer of the breast, 42 heavily pretreated postmenopausal patients received aminoglutethimide, 4 X 250 mg daily, with hydrocortisone or cortisone. Twenty-six received high doses of medroxyprogesterone before entering this study. There was no significant difference in patients' characteristics with or without medroxyprogesterone pretreatment. A comparison of patients with and without prior medroxyprogesterone shows a significant difference in the response rate to aminoglutethimide-hydrocortisone (4 vs 32%, P = 0.02). In patients pretreated with tamoxifen but not with medroxyprogesterone the response rate to aminoglutethimide was 36%. These results suggest that aminoglutethimide has a low activity in breast cancer patients previously exposed to medroxyprogesterone, an agent with glucocorticoid-like activity inducing adrenal suppression.
## 1847 Thirty-eight evaluable patients with metastatic breast cancer refractory to hormonal therapy and multiple chemotherapy regimens were treated with mitolactol at a dose of 130 mg/m2/day orally for 10 days every 6 weeks. Only one patient, with nodal and chest wall metastases, had a sustained complete regression; two patients had stable disease; and 35 patients had disease progression. The toxicity, which was primarily hematologic, was acceptable.
## 1852 High dose MAP (medroxyprogesterone acetate) and TAM (tamoxifen) were administered orally to 31 postmenopausal patients with advanced breast cancer. CR (complete remission) was achieved in 1 (3%) and PR (partial remission) in 11 (32%) patients for a median duration of 12 months. The treatment was well tolerated and in no instance was interrupted because of toxicity. This study does not support the hypothesis that the combination of MAP and TAM gives better results than the single agents.
## 1854 We compared two treatment regimens, transsphenoidal hypophysectomy and estrogen suppression with aminoglutethimide in women with metastatic breast carcinoma. Three of fourteen patients experienced partial objective tumor regression with a median duration of 4.6 months following hypophysectomy, whereas 10 of 21 women receiving aminoglutethimide responded (2 complete, 8 partial) with a median duration of 11.5 months. Side effects in the medical group were minimal while surgical complications included 2 cases of CSF rhinorrhea, one leading to meningitis and death. In patients receiving aminoglutethimide, urinary free cortisol and plasma dehydroepiandrosterone sulfate fell significantly as did plasma estrone and estradiol. In the hypophysectomy group, anterior-pituitary function testing postoperatively revealed adequate suppression of gonadotropin and prolactin secretion but incomplete inhibition of the ACTH-cortisol axis in 4 of 7 surgical patients studied. Five patients initially treated with hypophysectomy experienced a further reduction of plasma (and urinary) estrone and estradiol levels when given aminoglutethimide. We conclude that estrogen suppression therapy with aminoglutethimide is a feasible alternative to surgical hypophysectomy in providing endocrine suppression and palliation in advanced breast carcinoma.
## 1855
## trial
## 1842 1
## 1843 1
## 1847 1
## 1852 1
## 1854 0
## 1855 0
## 'data.frame': 558 obs. of 3 variables:
## $ title : chr "Randomized phase 3 trial of fluorouracil, epirubicin, and cyclophosphamide alone or followed by Paclitaxel for early breast can"| __truncated__ "Long-term endometrial effects in postmenopausal women with early breast cancer participating in the Intergroup Exemestane Study"| __truncated__ "Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant tamoxifen: Danish Breast Cancer Co"| __truncated__ "Adjuvant therapy of cancer via the cellular immune mechanism or fibrin by induced fibrinolysis and oral anticoagulants." ...
## $ abstract: chr "BACKGROUND: Taxanes are among the most active drugs for the treatment of metastatic breast cancer, and, as a consequence, they "| __truncated__ "BACKGROUND: The antiestrogen tamoxifen may have partial estrogen-like effects on the postmenopausal uterus. Aromatase inhibitor"| __truncated__ "BACKGROUND: Postmastectomy radiotherapy is associated with a lower locoregional recurrence rate and improved disease-free and o"| __truncated__ "The value of the oral anticoagulant warfarin sodium and fibrinolytic agents is discussed in relation to cancer surgery. A contr"| __truncated__ ...
## $ trial : int 1 0 0 0 1 1 0 0 0 0 ...
## - attr(*, "comment")= chr "glb_newent_df"
## title
## 1 Treatment of Hodgkin's disease and other cancers with 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU; NSC-409962).
## 2 Cell mediated immune status in malignancy--pretherapy and post-therapy assessment.
## 3 Neoadjuvant vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophosphamide in early nonresponsive breast cancer: phase III randomized GeparTrio trial.
## 5 Is first-line single-agent mitoxantrone in the treatment of high-risk metastatic breast cancer patients as effective as combination chemotherapy? No difference in survival but higher quality of life were found in a multicenter randomized trial.
## 6 Expression of Bcl-2 in node-negative breast cancer is associated with various prognostic factors, but does not predict response to one course of perioperative chemotherapy.
## 7 Combination docetaxel/cyclophosphamide in patients with advanced solid tumors.
## abstract
## 1
## 2 Twenty-eight cases of malignancies of different kinds were studied to assess T-cell activity and population before and after institution of therapy. Fifteen cases were diagnosed as non-metastasising squamous cell carcinoma of larynx, pharynx, laryngopharynx, hypopharynx and tonsils. Seven cases were non-metastasising infiltrating duct carcinoma of breast and 6 cases were non-Hodgkin's lymphoma (NHL). It was observed that 3 out of 15 cases (20%) of squamous cell carcinoma cases were Mantoux test (MT) negative with a T-cell population of less than 40%, 2 out of 7 cases (28.6%) of infiltrating duct carcinoma of breast were MT negative with a T-cell population of less than 40% and 3 out of 6 cases (50%) of NHL were MT negative with a T-cell population of less than 40%. The normal controls, consisting of apparently normal healthy adults, had a T-cell population of more than 40% and were all MT positive. The patients who showed a negative skin test and a T-cell population less than 40% were further subjected to assessment of T-cell population and activity after appropriate therapy, and clinical cure of the disease. It was observed that 2 out of 3 cases (66.66%) of squamous cell carcinomas, 2 out of 2 cases (100%) of adenocarcinomas and one out of 3 cases (33.33%) of NHL showed positive conversion with a T-cell population of more than 40%.
## 3 BACKGROUND: Among breast cancer patients, nonresponse to initial neoadjuvant chemotherapy is associated with unfavorable outcome. We compared the response of nonresponding patients who continued the same treatment with that of patients who switched to a well-tolerated non-cross-resistant regimen. METHODS: Previously untreated breast cancer patients received two 3-week cycles of docetaxel at 75 mg/m(2), doxorubicin at 50 mg/m(2), and cyclophosphamide at 500 mg/m(2) per day (TAC). Patients whose tumors did not decrease in size by at least 50% were randomly assigned to four additional cycles of TAC or to four cycles of vinorelbine at 25 mg/m(2) and capecitabine at 2000 mg/m(2) (NX). The outcome was sonographic response, defined as a reduction in the product of the two largest perpendicular diameters by at least 50%. A difference of 10% or less in the sonographic response qualified as noninferiority of the NX treatment. Pathological complete response was defined as no invasive or in situ residual tumor masses in the breast and lymph nodes. Toxic effects were assessed. All statistical tests were two-sided. RESULTS: Of 2090 patients enrolled in the GeparTrio study, 622 (29.8%) who did not respond to two initial cycles of TAC were randomly assigned to an additional four cycles of TAC (n = 321) or to four cycles of NX (n = 301). Sonographic response rate was 50.5% for the TAC arm and 51.2% for the NX arm. The difference of 0.7% (95% confidence interval = -7.1% to 8.5%) demonstrated noninferiority of NX (P = .008). Similar numbers of patients in both arms received breast-conserving surgery (184 [57.3%] in the TAC arm vs 180 [59.8%] in the NX arm) and had a pathological complete response (5.3% vs 6.0%). Fewer patients in the NX arm than in the TAC arm had hematologic toxic effects, mucositis, infections, and nail changes, but more had hand-foot syndrome and sensory neuropathy. CONCLUSION: Pathological complete responses to both regimens were marginal. Among patients who did not respond to the initial neoadjuvant TAC treatment, similar efficacy but better tolerability was observed by switching to NX than continuing with TAC.
## 5 BACKGROUND: To determine whether patients with high-risk metastatic breast cancer draw benefit from combination chemotherapy as first-line treatment. PATIENTS AND METHODS: A total of 260 women with measurable metastatic breast cancer fulfilling high-risk criteria, previously untreated with chemotherapy for their metastatic disease, were randomized to receive either mitoxantrone 12 mg/m(2) or the combination of fluorouracil 500 mg/m(2), epirubicin 50 mg/m(2) and cyclophosphamide 500 mg/m(2) (FEC) every 3 weeks. Treatment was continued until complete remission plus two cycles, or until disease progression. In the case of partial remission or stable disease, treatment was stopped after 12 cycles. Second-line treatment was vindesine, mitomycin and prednisolone. Gain from treatment was estimated using a modified Brunner's score composed of time to progression, patients' rating of the treatment benefit, alopecia, vomiting and performance status. RESULTS: After recruitment from 1992 to 1997 and observation from 1997 to 1999, the final evaluation showed that single-agent treatment with mitoxantrone does not differ significantly from combination treatment with FEC in terms of response, objective remission rate, remission duration, time to response, time to best response, time to progression or overall survival. There was, however, a significant difference in gain from treatment using a modified Brunner's score favoring the single-agent treatment arm. There was no evidence that any subgroup would fare better with combination treatment. CONCLUSIONS: No significant difference was detected between the treatment with mitoxantrone as a single agent and the combination of low-dose FEC in terms of response or survival; therefore, the imperative of the necessity of first-line combination chemotherapy for patients with high-risk metastatic breast cancer may be questioned. Since toxicity and quality of life score favored the single-agent mitoxantrone treatment arm, this treatment may be offered to patients preferring quality of life to a potential small prolongation of survival.
## 6 The aim of this study was to assess relationships between Bcl-2 expression, response to chemotherapy and a number of pathological and biological tumour parameters in premenopausal, lymph node-negative breast cancer patients. Expression of Bcl-2 was determined using immunohistochemistry on paraffin-embedded sections in a series of 441 premenopausal, lymph node-negative breast cancers of patients randomised to receive perioperative chemotherapy (5-fluorouracil, doxorubicin, cyclophosphamide) or no perioperative chemotherapy. Immunohistochemistry of Bcl-2 was evaluated by scoring both staining intensity (0-3) and number of positive cells (0-2). Using these scores tumours were grouped into categories 0-6. It was found that 9.2% of the tumours were completely negative (0), 17.2% weakly (1 + 2), 41.6% moderately (3 + 4) and 31.9% strongly positive (5 + 6) for Bcl-2. A positive correlation was found between high Bcl-2 expression and oestrogen (P < 0.001) and progesterone receptor positivity (P < 0.001) and low tumour grade (P < 0.001), whereas high Bcl-2 expression was negatively correlated with p53 (P < 0.001) and c-erb-B-2 positively (P < 0.001), high Ki-67 index (P < 0.001), mitotic index (P < 0.001) and large tumour size (P = 0.006). Patients with tumours expressing high levels of Bcl-2 (overall score 3-6) had a significantly better disease-free (P = 0.004) and overall (P = 0.009) survival. However, in a multivariate model this association no longer remained significant. There was a trend for an effect of adjuvant chemotherapy on disease-free survival both for patients with Bcl-2-positive (HR-0.61, 95% CI 0.35-1.06, P = 0.07) and negative (HR = 0.55, 95% CI 0.27-1.12, P = 0.09) breast tumours at a median follow-up of 49 months. The level of Bcl-2 expression does not seem to predict response to perioperative chemotherapy in premenopausal, lymph node-negative breast cancer patients. High levels of Bcl-2 are preferentially expressed in well-differentiated tumours and are associated with favourable prognosis. However, Bcl-2 expression is not an independent prognostic factor in this patient series.
## 7 Preclinical studies show that docetaxel (Taxotere) and cyclophosphamide (Cytoxan, Neosar) are synergistic against MA 13/C mammary adenocarcinoma. Both agents are highly active as monotherapy in a number of tumors, including metastatic breast cancer. Therefore, we performed a phase I dose-finding study to determine the maximum tolerated dose of this combination regimen in patients with advanced solid tumors. A total of 45 patients were enrolled and received cyclophosphamide followed by docetaxel, both administered as 1-hour intravenous infusions once every 3 weeks. The dose levels of cyclophosphamide/docetaxel were 600/60 mg/m2 (group 0), 600/75 mg/m2 (group I), 700/75 mg/m2 (group 2), 800/75 mg/m2 (group 3), 800/85 mg/m2 (group 4), 800/75 mg/m2 (group 5), and 800/85 mg/m2 (group 6). Patients with dose-limiting neutropenia in groups 5 and 6 received 300 micrograms of granulocyte colony-stimulating factor (G-CSF) (filgrastim [Neupogen]) support on days 2 through 9 during subsequent cycles of chemotherapy. All patients received premedication with 8 mg of dexamethasone twice daily for 5 days, beginning 1 day prior to chemotherapy. The dose-limiting toxicity was neutropenia fever. The recommended dose for phase II studies of cyclophosphamide/docetaxel is 700/75 mg/m2 in previously treated patients and 800/75 mg/m2 in previously untreated patients. G-CSF support did not allow for further dose escalation. Preliminary results from this phase I trial indicate that the combination of docetaxel and cyclophosphamide produced an objective response rate of 69% in 32 patients with metastatic breast cancer (including 3 patients who achieved complete responses).
## trial
## 1 1
## 2 0
## 3 1
## 5 1
## 6 0
## 7 1
## title
## 250 Assessment of residual tumour by FDG-PET: conventional imaging and clinical examination following primary chemotherapy of large and locally advanced breast cancer.
## 253 HER2, TOP2A, and TIMP-1 and responsiveness to adjuvant anthracycline-containing chemotherapy in high-risk breast cancer patients.
## 510 Phase II, randomized trial to compare anastrozole combined with gefitinib or placebo in postmenopausal women with hormone receptor-positive metastatic breast cancer.
## 541 Epidermal growth factor receptor and vascular endothelial growth factor receptor 2 are specific biomarkers in triple-negative breast cancer. Results from a controlled randomized trial with long-term follow-up.
## 820 Estrogen-receptor status and outcomes of modern chemotherapy for patients with node-positive breast cancer.
## 878 Postoperative chemo-endocrine treatment with mitomycin C, tamoxifen, and UFT is effective for patients with premenopausal estrogen receptor-positive stage II breast cancer. Nishinihon Cooperative Study Group of Adjuvant Therapy for Breast Cancer.
## abstract
## 250 BACKGROUND: The aim of this was to evaluate FDG-PET (2-(fluorine-18)-fluoro-2-deoxy-D-glucose positron emission tomography) for assessment of residual tumour after primary chemotherapy of large and locally advanced breast cancer in comparison with conventional imaging modalities.METHODS: In a prospective multicentre trial, 99 patients underwent one or more breast imaging modalities before surgery in addition to clinical examination, namely, FDG-PET (n=89), mammography (n=47), ultrasound (n=46), and magnetic resonance imaging (MRI) (n=46). The presence of residual tumour by conventional imaging, dichotomised as positive or negative, and the level of FDG uptake (standardised uptake values, SUV) were compared with histopathology, which served as the reference standard. Patients with no residual tumour or only small microscopic foci of residual tumour were classified as having minimal residual disease and those with extensive microscopic and macroscopic residual tumour tissue were classified as having gross residual disease.RESULTS: By applying a threshold SUV of 2.0, the sensitivity of FDG-PET for residual tumour was 32.9% (specificity, 87.5%) and increased to 57.5% (specificity, 62.5%) at a threshold SUV of 1.5. Conventional imaging modalities were more sensitive in identifying residual tumour, but had a low corresponding specificity; sensitivity and specificity were as follows: MRI 97.6 and 40.0%, mammography 92.5 and 57.1%, ultrasound 92.0 and 37.5%, respectively. Breast MRI provided the highest accuracy (91.3%), whereas FDG-PET had the lowest accuracy (42.7%).CONCLUSIONS: FDG-PET does not provide an accurate assessment of residual tumour after primary chemotherapy of breast cancer. Magnetic resonance imaging offers the highest sensitivity, but all imaging modalities have distinct limitations in the assessment of residual tumour tissue when compared with histopathology.
## 253 PURPOSE To evaluate whether the combination of HER2 with TIMP-1 (HT) or TOP2A with TIMP-1 (2T) more accurately identifies patients who benefit from cyclophosphamide, epirubicin, and fluorouracil (CEF) compared with cyclophosphamide, methotrexate, and fluorouracil (CMF) than these markers do when analyzed individually. PATIENTS AND METHODS The Danish Breast Cancer Cooperative Group (DBCG) 89D trial randomly assigned 980 high-risk Danish breast cancer patients to CMF or CEF. Archival tumor tissue was analyzed TIMP-1, and HER2-negative and TIMP-1 immunoreactive tumors were classified as HT nonresponsive and otherwise HT responsive. Similarly, the 2T panel was constructed by combining TOP2A and TIMP-1; tumors with normal TOP2A status and TIMP-1 immunoreactivity were classified as 2T-nonresponsive and otherwise 2T-responsive. Results In total, 623 tumors were available for analysis, of which 154 lacked TIMP-1 immunoreactivity, 188 were HER2 positive, and 139 had a TOP2A aberration. HT status was a statistically significant predictor of benefit from CEF compared with CMF (P(interaction) = .036 for invasive disease-free survival [IDFS] and .047 for overall survival [OS]). The 269 (43%) patients with a 2T-responsive profile had a significant reduction in IDFS events (adjusted hazard ratio, 0.48; 95% CI, 0.34 to 0.69; P < .001) and OS events (adjusted hazard ratio, 0.54; 95% CI, 0.38 to 0.77; P < .001). 2T status was a highly significant predictor of benefit from CEF compared with CMF (P(interaction) < .0001 for IDFS and .004 for OS). CONCLUSION The 2T profile is a more accurate predictor of incremental benefit from anthracycline-containing chemotherapy than HER2, TIMP-1, or TOP2A individually, and compared with these, 2T classifies a larger proportion of patients as sensitive to anthracyclines.
## 510 PURPOSE: This phase II randomized trial evaluated the efficacy and tolerability of anastrozole combined with gefitinib or anastrozole with placebo in women with hormone receptor-positive metastatic breast cancer (MBC). EXPERIMENTAL DESIGN: Postmenopausal women with hormone receptor-positive measurable or evaluable MBC who had not received prior endocrine therapy for this disease stage or who developed metastatic disease during/after adjuvant tamoxifen were eligible. The primary response variable was progression-free survival (PFS) and secondary response variables included clinical benefit rate, objective response rate, overall survival, safety and tolerability, and pharmacokinetics. Tumor biomarker evaluation was an exploratory objective. RESULTS: Forty-three patients were randomized to anastrozole plus gefitinib and 50 patients were randomized to anastrozole plus placebo of a planned total of 174 patients (enrollment was prematurely discontinued due to slow recruitment). PFS for patients receiving the combination of anastrozole and gefitinib was longer than for patients receiving anastrozole plus placebo [hazard ratio (gefitinib/placebo), 0.55; 95% confidence interval, 0.32-0.94; median PFS, 14.7 versus 8.4 months]. The clinical benefit rate was 49% versus 34%, and the objective response rate was 2% versus 12% with anastrozole plus gefitinib and anastrozole plus placebo, respectively. No evidence of interaction between baseline biomarker levels and relative treatment effect was found. No unexpected adverse events were observed. CONCLUSION: This small randomized study showed that anastrozole in combination with gefitinib is associated with a marked advantage in PFS compared with anastrozole plus placebo, and that the combination was tolerated in postmenopausal women with hormone receptor-positive MBC. Further investigation of epidermal growth factor receptor inhibition in combination with endocrine therapy may be warranted.
## 541 Triple-negative breast cancer (TNB) has poor prognosis and moreover patients with TNB do not benefit from established targeted drugs with endocrine therapy or trastuzumab. The aim of the study was to analyze the prevalence of candidate biomarkers in tumors from patients with TNB. Tissue microarrays were prepared from primary tumors from premenopausal breast cancer patients (500/564) randomized to adjuvant tamoxifen or no adjuvant treatment. Immunohistochemical (IHC) staining included ER, PR, HER2, epidermal receptor growth factor (EGFR), vascular endothelial growth factor A (VEGF-A), and vascular endothelial growth factor receptor 2 (VEGFR2). EGFR and HER2 gene copy number was defined by fluorescence in situ hybridization (FISH). All patients were included in the descriptive analysis, but only untreated patients in the survival analysis. TNB was diagnosed in 96 patients and correlated significantly to low age, Nottingham histological grade (NHG) III, high Ki67-index, T2 tumors, node negativity, EGFR positivity, increased EGFR gene copy number and high VEGFR2 expression. TNB was an independent prognostic factor for decreased 5-year breast cancer specific survival (BCSS) (HR 2.0 (95% CI 1.1-3.6), P = 0.01), but not for 10-year BCSS. High VEGFR2 expression was significantly correlated to decreased BCSS in TNB patients. TNB was associated with decreased BCSS and clinicopathological characteristics of an aggressive tumor type. High VEGFR2 expression, EGFR expression, and EGFR gene copy number were significantly correlated to TNB, supporting their role as putative candidate biomarkers for selection of targeted therapy in TNB.
## 820 CONTEXT: Breast cancer estrogen-receptor (ER) status is useful in predicting benefit from endocrine therapy. It may also help predict which patients benefit from advances in adjuvant chemotherapy. OBJECTIVE: To compare differences in benefits from adjuvant chemotherapy achieved by patients with ER-negative vs ER-positive tumors. DESIGN, SETTING, AND PATIENTS: Trial data from the Cancer and Leukemia Group B and US Breast Cancer Intergroup analyzed; patient outcomes by ER status compared using hazards over time and multivariate models. Randomized trials comparing (1): 3 regimens of cyclophosphamide, doxorubicin, and fluorouracil (January 1985 to April 1991); (2) 3 doses of doxorubicin concurrent with cyclophosphamide, with or without subsequent paclitaxel (May 1994 to April 1997); (3) sequential doxorubicin, paclitaxel, and cyclophosphamide with concurrent doxorubicin and cyclophosphamide followed by paclitaxel, and also 3-week vs 2-week cycles (September 1997 to March 1999). A total of 6644 node-positive breast cancer patients received adjuvant treatment. MAIN OUTCOME MEASURES: Disease-free and overall survival. RESULTS: For ER-negative tumors, chemotherapy improvements reduced the relative risk of recurrence by 21%, 25%, and 23% in the 3 studies, respectively, and 55% comparing the lowest dose in the first study with biweekly cycles in the third study. Corresponding relative risk reductions for ER-positive tumors treated with tamoxifen were 9%, 12%, and 8% in the 3 studies, and 26% overall. The overall mortality rate reductions associated with chemotherapy improvements were 55% and 23% among ER-negative and ER-positive patients, respectively. All individual ER-negative comparisons and no ER-positive comparisons were statistically significant. Absolute benefits due to chemotherapy were greater for patients with ER-negative compared with ER-positive tumors: 22.8% more ER-negative patients survived to 5 years disease-free if receiving chemotherapy vs 7.0% for ER-positive patients; corresponding improvements for overall survival were 16.7% vs 4.0%. CONCLUSION: Among patients with node-positive tumors, ER-negative breast cancer, biweekly doxorubicin/cyclophosphamide plus paclitaxel lowers the rate of recurrence and death by more than 50% in comparison with low-dose cyclophosphamide, doxorubicin, and fluorouracil as used in the first study.
## 878 The effectiveness of combining mitomycin C (MMC), tamoxifen (TAM), and 1-(2-tetrahydrofuryl)-5-fluorouracil (tegafur) was evident in patients with estrogen receptor-positive (ER+) breast cancers. UFT, an oral preparation of tegafur and uracil at a molar ratio of 1:4, was reported to have higher antitumor effects than tegafur alone for patients with breast cancer. Therefore, the combined chemotherapy of MMC, TAM and UFT may possibly be effective for breast cancer. From 1988 to 1991. we studied the effects of postoperative adjuvant therapy for Japanese women with stage 11 breast cancer, all seen at 71 institutions in western areas of Japan. Five hundred and ninety four patients with stage II primary breast cancer who had undergone curative surgery, including total mastectomy and axillary lymph node dissection, were enrolled. On the day of surgery, each patient was given 13 mg/m2 of MMC intravenously. Patients with ER+ tumors were then assigned to group A or group B. Group A received 30 mg/day of TAM given orally from postoperative 2 weeks, for 2 years. Group B was additionally given an oral dose of 300 mg/day of UFT for 2 years, given concomitantly with 30 mg/day of TAM. Patients with ER- tumors were assigned to group C or group D. Group C were prescribed 300 mg/day of UFT, orally, from postoperative 2 weeks for 2 years, and group D were additionally given an oral dose of 30 mg/day of TAM together with 300 mg/day of UFT. There were no differences among the groups regarding prognostic factors or doses of MMC and TAM in ER+ patients and MMC and UFT in ER- patients. Toxicity rates for leukopenia, anorexia, and nausea/vomiting were higher in group B than in group A patients. There were no statistical differences in the overall survival and disease-free survival times between groups A and B, or groups C and D, for all eligible cases. In a retrospective subgroup analysis using Bonferroni's adjustments, the additional effect of UFT on the combined treatment of MMC and TAM lengthened the disease-free survival time for patients with premenopausal ER+ cancers (corrected P value by Bonferroni's adjustments <0.05). Multivariate analysis showed that effects of the combined treatment of MMC, TAM, and UFT was significantly related to the menopausal status (P<0.01). Our findings show that postoperative ingestion of MMC, TAM, and UFT was effective for patients with premenopausal ER+ stage II breast cancer.
## trial
## 250 0
## 253 0
## 510 1
## 541 0
## 820 0
## 878 1
## title
## 1853 Sequential methotrexate plus 5-FU in advanced breast and colorectal cancers: a phase II study.
## 1856 Clinical trial of 2-Br- -ergocryptine (CB154) in advanced breast cancer.
## 1857 Five-drug therapy for advanced breast cancer: a phase I study.
## 1858 A controlled clinical trial to compare transethidal hypophysectomy with yttrium implant of the pituitary in the treatment of advanced breast cancer.
## 1859 A clinical trial to compare early and late pituitary ablation in advanced cancer of the breast.
## 1860 A clinical trial of tri-iodothyronine as a hormone potentiator in advanced breast cancer.
## abstract
## 1853 Fifty-two patients with advanced breast or colorectal cancer have been treated with methotrexate (MTX) (50 mg/m2) followed 6 hours later by 5-FU (600 mg/m2). The mean serum MTX level immediately prior to 5-FU administration was 1.37 mumols/L (1.37 X 10(-6) M). Of 29 evaluable patients with breast cancer (six of whom had received prior chemotherapy), one achieved a complete response and five achieved a partial response (total response rate, 21%). Among 16 evaluable patients with colorectal cancer (three of whom had received prior chemotherapy), there were no objective responses. Although hematologic toxicity was generally mild, mucositis occurred in 20 patients (severe in three), and at least one early death was attributable to toxicity. These results indicate that at doses of MTX and 5-FU which are in general use in combination chemotherapy for breast cancer, sequential treatment does not have a therapeutic advantage. In colorectal cancer, the same combination is inactive.
## 1856
## 1857
## 1858
## 1859
## 1860
## trial
## 1853 1
## 1856 0
## 1857 1
## 1858 0
## 1859 0
## 1860 0
## 'data.frame': 1302 obs. of 3 variables:
## $ title : chr "Treatment of Hodgkin's disease and other cancers with 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU; NSC-409962)." "Cell mediated immune status in malignancy--pretherapy and post-therapy assessment." "Neoadjuvant vinorelbine-capecitabine versus docetaxel-doxorubicin-cyclophosphamide in early nonresponsive breast cancer: phase "| __truncated__ "Is first-line single-agent mitoxantrone in the treatment of high-risk metastatic breast cancer patients as effective as combina"| __truncated__ ...
## $ abstract: chr "" "Twenty-eight cases of malignancies of different kinds were studied to assess T-cell activity and population before and after in"| __truncated__ "BACKGROUND: Among breast cancer patients, nonresponse to initial neoadjuvant chemotherapy is associated with unfavorable outcom"| __truncated__ "BACKGROUND: To determine whether patients with high-risk metastatic breast cancer draw benefit from combination chemotherapy as"| __truncated__ ...
## $ trial : int 1 0 1 1 0 1 0 0 1 0 ...
## - attr(*, "comment")= chr "glb_trnent_df"if (!is.null(glb_max_trnent_obs)) {
if (nrow(glb_trnent_df) > glb_max_trnent_obs) {
warning("glb_trnent_df restricted to glb_max_trnent_obs: ",
format(glb_max_trnent_obs, big.mark=","))
org_entity_df <- glb_trnent_df
glb_trnent_df <- org_entity_df[split <-
sample.split(org_entity_df[, glb_rsp_var_raw],
SplitRatio=glb_max_trnent_obs), ]
org_entity_df <- NULL
}
# if (nrow(glb_newent_df) > glb_max_obs) {
# warning("glb_newent_df restricted to glb_max_obs: ", format(glb_max_obs, big.mark=","))
# org_newent_df <- glb_newent_df
# glb_newent_df <- org_newent_df[split <-
# sample.split(org_newent_df[, glb_rsp_var_raw], SplitRatio=glb_max_obs), ]
# org_newent_df <- NULL
# }
}
if (nrow(glb_trnent_df) == nrow(glb_entity_df))
warning("glb_trnent_df same as glb_entity_df")
if (nrow(glb_newent_df) == nrow(glb_entity_df))
warning("glb_newent_df same as glb_entity_df")
if (length(glb_drop_vars) > 0) {
warning("dropping vars: ", paste0(glb_drop_vars, collapse=", "))
glb_entity_df <- glb_entity_df[, setdiff(names(glb_entity_df), glb_drop_vars)]
glb_trnent_df <- glb_trnent_df[, setdiff(names(glb_trnent_df), glb_drop_vars)]
glb_newent_df <- glb_newent_df[, setdiff(names(glb_newent_df), glb_drop_vars)]
}
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="cleanse_data",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed import_data 1 0 0.002
## elapsed1 cleanse_data 2 0 1.422Step 2: cleanse data
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="inspectORexplore.data",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed1 cleanse_data 2 0 1.422
## elapsed2 inspectORexplore.data 2 1 1.504Step 2.1: inspect/explore data
#print(str(glb_trnent_df))
#View(glb_trnent_df)
# List info gathered for various columns
# <col_name>: <description>; <notes>
# Create new features that help diagnostics
# Create factors of string variables
str_vars <- sapply(1:length(names(glb_trnent_df)),
function(col) ifelse(class(glb_trnent_df[, names(glb_trnent_df)[col]]) == "character",
names(glb_trnent_df)[col], ""))
if (length(str_vars <- setdiff(str_vars[str_vars != ""],
glb_exclude_vars_as_features)) > 0) {
warning("Creating factors of string variables:", paste0(str_vars, collapse=", "))
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, str_vars)
for (var in str_vars) {
glb_entity_df[, paste0(var, ".fctr")] <- factor(glb_entity_df[, var],
as.factor(unique(glb_entity_df[, var])))
glb_trnent_df[, paste0(var, ".fctr")] <- factor(glb_trnent_df[, var],
as.factor(unique(glb_entity_df[, var])))
glb_newent_df[, paste0(var, ".fctr")] <- factor(glb_newent_df[, var],
as.factor(unique(glb_entity_df[, var])))
}
}
# Convert factors to dummy variables
# Build splines require(splines); bsBasis <- bs(training$age, df=3)
add_new_diag_feats <- function(obs_df, ref_df=glb_entity_df) {
require(plyr)
obs_df <- mutate(obs_df,
# <col_name>.NA=is.na(<col_name>),
# <col_name>.fctr=factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# <col_name>.fctr=relevel(factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# "<ref_val>"),
# <col2_name>.fctr=relevel(factor(ifelse(<col1_name> == <val>, "<oth_val>", "<ref_val>")),
# as.factor(c("R", "<ref_val>")),
# ref="<ref_val>"),
# This doesn't work - use sapply instead
# <col_name>.fctr_num=grep(<col_name>, levels(<col_name>.fctr)),
#
# Date.my=as.Date(strptime(Date, "%m/%d/%y %H:%M")),
# Year=year(Date.my),
# Month=months(Date.my),
# Weekday=weekdays(Date.my)
# <col_name>.log=log(<col.name>),
# <col_name>=<table>[as.character(<col2_name>)],
# <col_name>=as.numeric(<col2_name>),
.rnorm=rnorm(n=nrow(obs_df))
)
# If levels of a factor are different across obs_df & glb_newent_df; predict.glm fails
# Transformations not handled by mutate
# obs_df$<col_name>.fctr.num <- sapply(1:nrow(obs_df),
# function(row_ix) grep(obs_df[row_ix, "<col_name>"],
# levels(obs_df[row_ix, "<col_name>.fctr"])))
print(summary(obs_df))
print(sapply(names(obs_df), function(col) sum(is.na(obs_df[, col]))))
return(obs_df)
}
glb_entity_df <- add_new_diag_feats(glb_entity_df)## Loading required package: plyr## title abstract trial .rnorm
## Length:1860 Length:1860 Min. :0.0000 Min. :-2.87472
## Class :character Class :character 1st Qu.:0.0000 1st Qu.:-0.63414
## Mode :character Mode :character Median :0.0000 Median :-0.01097
## Mean :0.4392 Mean : 0.02616
## 3rd Qu.:1.0000 3rd Qu.: 0.72265
## Max. :1.0000 Max. : 3.48994
## title abstract trial .rnorm
## 0 0 0 0glb_trnent_df <- add_new_diag_feats(glb_trnent_df)## title abstract trial
## Length:1302 Length:1302 Min. :0.0000
## Class :character Class :character 1st Qu.:0.0000
## Mode :character Mode :character Median :0.0000
## Mean :0.4393
## 3rd Qu.:1.0000
## Max. :1.0000
## .rnorm
## Min. :-3.323839
## 1st Qu.:-0.665678
## Median :-0.025640
## Mean : 0.002532
## 3rd Qu.: 0.703387
## Max. : 3.197964
## title abstract trial .rnorm
## 0 0 0 0glb_newent_df <- add_new_diag_feats(glb_newent_df)## title abstract trial .rnorm
## Length:558 Length:558 Min. :0.0000 Min. :-3.37419
## Class :character Class :character 1st Qu.:0.0000 1st Qu.:-0.64421
## Mode :character Mode :character Median :0.0000 Median : 0.03015
## Mean :0.4391 Mean : 0.01893
## 3rd Qu.:1.0000 3rd Qu.: 0.68957
## Max. :1.0000 Max. : 3.16282
## title abstract trial .rnorm
## 0 0 0 0# Histogram of predictor in glb_trnent_df & glb_newent_df
plot_df <- rbind(cbind(glb_trnent_df[, glb_rsp_var_raw, FALSE], data.frame(.data="Training")),
cbind(glb_trnent_df[, glb_rsp_var_raw, FALSE], data.frame(.data="New")))
print(myplot_histogram(plot_df, glb_rsp_var_raw) + facet_wrap(~ .data))## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
# used later in encode.retype.data chunk
glb_display_class_dstrb <- function(var) {
plot_df <- rbind(cbind(glb_trnent_df[, var, FALSE],
data.frame(.data="Training")),
cbind(glb_trnent_df[, var, FALSE],
data.frame(.data="New")))
xtab_df <- mycreate_xtab(plot_df, c(".data", var))
rownames(xtab_df) <- xtab_df$.data
xtab_df <- subset(xtab_df, select=-.data)
print(xtab_df / rowSums(xtab_df))
}
if (glb_is_classification) glb_display_class_dstrb(glb_rsp_var_raw)## Loading required package: reshape2## trial.0 trial.1
## New 0.5606759 0.4393241
## Training 0.5606759 0.4393241# Check for duplicates in glb_id_vars
if (length(glb_id_vars) > 0) {
id_vars_dups_df <- subset(id_vars_df <-
mycreate_tbl_df(glb_entity_df[, glb_id_vars, FALSE], glb_id_vars),
.freq > 1)
} else {
tmp_entity_df <- glb_entity_df
tmp_entity_df$.rownames <- rownames(tmp_entity_df)
id_vars_dups_df <- subset(id_vars_df <-
mycreate_tbl_df(tmp_entity_df[, ".rownames", FALSE], ".rownames"),
.freq > 1)
}
if (nrow(id_vars_dups_df) > 0) {
warning("Duplicates found in glb_id_vars data:", nrow(id_vars_dups_df))
myprint_df(id_vars_dups_df)
} else {
# glb_id_vars are unique across obs in both glb_<>_df
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, glb_id_vars)
}
#pairs(subset(glb_trnent_df, select=-c(col_symbol)))
# Check for glb_newent_df & glb_trnent_df features range mismatches
# Other diagnostics:
# print(subset(glb_trnent_df, <col1_name> == max(glb_trnent_df$<col1_name>, na.rm=TRUE) &
# <col2_name> <= mean(glb_trnent_df$<col1_name>, na.rm=TRUE)))
# print(glb_trnent_df[which.max(glb_trnent_df$<col_name>),])
# print(<col_name>_freq_glb_trnent_df <- mycreate_tbl_df(glb_trnent_df, "<col_name>"))
# print(which.min(table(glb_trnent_df$<col_name>)))
# print(which.max(table(glb_trnent_df$<col_name>)))
# print(which.max(table(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>)[, 2]))
# print(table(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>))
# print(table(is.na(glb_trnent_df$<col1_name>), glb_trnent_df$<col2_name>))
# print(table(sign(glb_trnent_df$<col1_name>), glb_trnent_df$<col2_name>))
# print(mycreate_xtab(glb_trnent_df, <col1_name>))
# print(mycreate_xtab(glb_trnent_df, c(<col1_name>, <col2_name>)))
# print(<col1_name>_<col2_name>_xtab_glb_trnent_df <-
# mycreate_xtab(glb_trnent_df, c("<col1_name>", "<col2_name>")))
# <col1_name>_<col2_name>_xtab_glb_trnent_df[is.na(<col1_name>_<col2_name>_xtab_glb_trnent_df)] <- 0
# print(<col1_name>_<col2_name>_xtab_glb_trnent_df <-
# mutate(<col1_name>_<col2_name>_xtab_glb_trnent_df,
# <col3_name>=(<col1_name> * 1.0) / (<col1_name> + <col2_name>)))
# print(<col2_name>_min_entity_arr <-
# sort(tapply(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>, min, na.rm=TRUE)))
# print(<col1_name>_na_by_<col2_name>_arr <-
# sort(tapply(glb_trnent_df$<col1_name>.NA, glb_trnent_df$<col2_name>, mean, na.rm=TRUE)))
# Other plots:
# print(myplot_box(df=glb_trnent_df, ycol_names="<col1_name>"))
# print(myplot_box(df=glb_trnent_df, ycol_names="<col1_name>", xcol_name="<col2_name>"))
# print(myplot_line(subset(glb_trnent_df, Symbol %in% c("KO", "PG")),
# "Date.my", "StockPrice", facet_row_colnames="Symbol") +
# geom_vline(xintercept=as.numeric(as.Date("2003-03-01"))) +
# geom_vline(xintercept=as.numeric(as.Date("1983-01-01")))
# )
# print(myplot_scatter(glb_entity_df, "<col1_name>", "<col2_name>", smooth=TRUE))
# print(myplot_scatter(glb_entity_df, "<col1_name>", "<col2_name>", colorcol_name="<Pred.fctr>") +
# geom_point(data=subset(glb_entity_df, <condition>),
# mapping=aes(x=<x_var>, y=<y_var>), color="red", shape=4, size=5))
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="manage_missing_data",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed2 inspectORexplore.data 2 1 1.504
## elapsed3 manage_missing_data 2 2 2.694Step 2.2: manage missing data
# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
# glb_trnent_df <- na.omit(glb_trnent_df)
# glb_newent_df <- na.omit(glb_newent_df)
# df[is.na(df)] <- 0
# Not refactored into mydsutils.R since glb_*_df might be reassigned
glb_impute_missing_data <- function(entity_df, newent_df) {
if (!glb_is_separate_newent_dataset) {
# Combine entity & newent
union_df <- rbind(mutate(entity_df, .src = "entity"),
mutate(newent_df, .src = "newent"))
union_imputed_df <- union_df[, setdiff(setdiff(names(entity_df),
glb_rsp_var),
glb_exclude_vars_as_features)]
print(summary(union_imputed_df))
require(mice)
set.seed(glb_mice_complete.seed)
union_imputed_df <- complete(mice(union_imputed_df))
print(summary(union_imputed_df))
union_df[, names(union_imputed_df)] <- union_imputed_df[, names(union_imputed_df)]
print(summary(union_df))
# union_df$.rownames <- rownames(union_df)
# union_df <- orderBy(~.rownames, union_df)
#
# imp_entity_df <- myimport_data(
# url="<imputed_trnng_url>",
# comment="imp_entity_df", force_header=TRUE, print_diagn=TRUE)
# print(all.equal(subset(union_df, select=-c(.src, .rownames, .rnorm)),
# imp_entity_df))
# Partition again
glb_trnent_df <<- subset(union_df, .src == "entity", select=-c(.src, .rownames))
comment(glb_trnent_df) <- "entity_df"
glb_newent_df <<- subset(union_df, .src == "newent", select=-c(.src, .rownames))
comment(glb_newent_df) <- "newent_df"
# Generate summaries
print(summary(entity_df))
print(sapply(names(entity_df), function(col) sum(is.na(entity_df[, col]))))
print(summary(newent_df))
print(sapply(names(newent_df), function(col) sum(is.na(newent_df[, col]))))
} else stop("Not implemented yet")
}
if (glb_impute_na_data) {
if ((sum(sapply(names(glb_trnent_df),
function(col) sum(is.na(glb_trnent_df[, col])))) > 0) |
(sum(sapply(names(glb_newent_df),
function(col) sum(is.na(glb_newent_df[, col])))) > 0))
glb_impute_missing_data(glb_trnent_df, glb_newent_df)
}
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="encodeORretype.data",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed3 manage_missing_data 2 2 2.694
## elapsed4 encodeORretype.data 2 3 2.973Step 2.3: encode/retype data
# map_<col_name>_df <- myimport_data(
# url="<map_url>",
# comment="map_<col_name>_df", print_diagn=TRUE)
# map_<col_name>_df <- read.csv(paste0(getwd(), "/data/<file_name>.csv"), strip.white=TRUE)
# glb_trnent_df <- mymap_codes(glb_trnent_df, "<from_col_name>", "<to_col_name>",
# map_<to_col_name>_df, map_join_col_name="<map_join_col_name>",
# map_tgt_col_name="<to_col_name>")
# glb_newent_df <- mymap_codes(glb_newent_df, "<from_col_name>", "<to_col_name>",
# map_<to_col_name>_df, map_join_col_name="<map_join_col_name>",
# map_tgt_col_name="<to_col_name>")
# glb_trnent_df$<col_name>.fctr <- factor(glb_trnent_df$<col_name>,
# as.factor(union(glb_trnent_df$<col_name>, glb_newent_df$<col_name>)))
# glb_newent_df$<col_name>.fctr <- factor(glb_newent_df$<col_name>,
# as.factor(union(glb_trnent_df$<col_name>, glb_newent_df$<col_name>)))
if (!is.null(glb_map_rsp_raw_to_var)) {
glb_entity_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_entity_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_entity_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
glb_trnent_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_trnent_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_trnent_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
glb_newent_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_newent_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_newent_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
if (glb_is_classification) glb_display_class_dstrb(glb_rsp_var)
}## Loading required package: sqldf
## Loading required package: gsubfn
## Loading required package: proto
## Loading required package: RSQLite
## Loading required package: DBI
## Loading required package: tcltk## trial trial.fctr .n
## 1 0 N 1043
## 2 1 Y 817
## trial trial.fctr .n
## 1 0 N 730
## 2 1 Y 572
## trial trial.fctr .n
## 1 0 N 313
## 2 1 Y 245
## trial.fctr.N trial.fctr.Y
## New 0.5606759 0.4393241
## Training 0.5606759 0.4393241glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="extract.features",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed4 encodeORretype.data 2 3 2.973
## elapsed5 extract.features 3 0 6.239Step 3: extract features
#```{r extract_features, cache=FALSE, eval=glb_is_textual}
# Create new features that help prediction
# <col_name>.lag.2 <- lag(zoo(glb_trnent_df$<col_name>), -2, na.pad=TRUE)
# glb_trnent_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
# <col_name>.lag.2 <- lag(zoo(glb_newent_df$<col_name>), -2, na.pad=TRUE)
# glb_newent_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
#
# glb_newent_df[1, "<col_name>.lag.2"] <- glb_trnent_df[nrow(glb_trnent_df) - 1,
# "<col_name>"]
# glb_newent_df[2, "<col_name>.lag.2"] <- glb_trnent_df[nrow(glb_trnent_df),
# "<col_name>"]
# glb_entity_df <- mutate(glb_entity_df,
# A.has.http=ifelse(grepl("http",Added,fixed=TRUE), 1, 0)
# )
#
# glb_trnent_df <- mutate(glb_trnent_df,
# )
#
# glb_newent_df <- mutate(glb_newent_df,
# )
if (glb_is_textual) {
require(tm)
glb_corpus_lst <- list(); glb_full_DTM_lst <- list(); glb_sprs_DTM_lst <- list();
for (txt_var in glb_txt_vars) {
print(sprintf("Building corpus for %s...", txt_var))
txt_corpus <- Corpus(VectorSource(glb_entity_df[, txt_var]))
txt_corpus <- tm_map(txt_corpus, tolower)
txt_corpus <- tm_map(txt_corpus, PlainTextDocument)
txt_corpus <- tm_map(txt_corpus, removePunctuation)
txt_corpus <- tm_map(txt_corpus, removeWords,
c(glb_append_stop_words, stopwords("english")))
txt_corpus <- tm_map(txt_corpus, stemDocument)
full_freqs_DTM <- DocumentTermMatrix(txt_corpus)
print(" Full freqs:"); print(full_freqs_DTM)
full_freqs_vctr <- colSums(as.matrix(full_freqs_DTM))
names(full_freqs_vctr) <- dimnames(full_freqs_DTM)[[2]]
full_freqs_df <- as.data.frame(full_freqs_vctr)
names(full_freqs_df) <- "freq.full"
full_freqs_df$term <- rownames(full_freqs_df)
full_freqs_df <- orderBy(~ -freq.full, full_freqs_df)
sprs_freqs_DTM <- removeSparseTerms(full_freqs_DTM, glb_sprs_threshold)
print(" Sparse freqs:"); print(sprs_freqs_DTM)
sprs_freqs_vctr <- colSums(as.matrix(sprs_freqs_DTM))
names(sprs_freqs_vctr) <- dimnames(sprs_freqs_DTM)[[2]]
sprs_freqs_df <- as.data.frame(sprs_freqs_vctr)
names(sprs_freqs_df) <- "freq.sprs"
sprs_freqs_df$term <- rownames(sprs_freqs_df)
sprs_freqs_df <- orderBy(~ -freq.sprs, sprs_freqs_df)
terms_freqs_df <- merge(full_freqs_df, sprs_freqs_df, all.x=TRUE)
melt_freqs_df <- orderBy(~ -value, melt(terms_freqs_df, id.var="term"))
print(ggplot(melt_freqs_df, aes(value, color=variable)) + stat_ecdf() +
geom_hline(yintercept=glb_sprs_threshold, linetype = "dotted"))
print(myplot_hbar(head(melt_freqs_df, 20), "term", "value",
colorcol_name="variable"))
melt_freqs_df <- orderBy(~ -value,
melt(subset(terms_freqs_df, is.na(freq.sprs)), id.var="term"))
print(myplot_hbar(head(melt_freqs_df, 10), "term", "value",
colorcol_name="variable"))
glb_corpus_lst[[txt_var]] <- txt_corpus
glb_full_DTM_lst[[txt_var]] <- full_freqs_DTM
glb_sprs_DTM_lst[[txt_var]] <- sprs_freqs_DTM
}
# Create txt features
if ((length(glb_txt_vars) > 1) &&
(length(unique(pfxs <- sapply(glb_txt_vars,
function(txt) toupper(substr(txt, 1, 1))))) < length(glb_txt_vars)))
stop("Prefixes for corpus freq terms not unique: ", pfxs)
for (txt_var in glb_txt_vars) {
print(sprintf("Binding DTM for %s...", txt_var))
txt_X_df <- as.data.frame(as.matrix(glb_sprs_DTM_lst[[txt_var]]))
colnames(txt_X_df) <- paste(toupper(substr(txt_var, 1, 1)), ".",
make.names(colnames(txt_X_df)), sep="")
rownames(txt_X_df) <- rownames(glb_entity_df) # warning otherwise
glb_entity_df <- cbind(glb_entity_df, txt_X_df)
# Create <txt_var>.has.http
glb_entity_df[, paste(toupper(substr(txt_var, 1, 1)), ".has.http", sep="")] <-
sapply(1:nrow(glb_entity_df),
function(row_ix) ifelse(grepl("http", glb_entity_df[row_ix, txt_var], fixed=TRUE),
1, 0))
# Create <txt_var>.num.chars
glb_entity_df[, paste(toupper(substr(txt_var, 1, 1)), ".num.chars", sep="")] <-
sapply(1:nrow(glb_entity_df),
function(row_ix) nchar(glb_entity_df[row_ix, txt_var]))
# Create <txt_var>.num.words & .num.words.unq
glb_entity_df[, paste(toupper(substr(txt_var, 1, 1)), ".num.words", sep="")] <-
rowSums(as.matrix(glb_full_DTM_lst[[txt_var]]))
glb_entity_df[, paste(toupper(substr(txt_var, 1, 1)), ".num.words.unq", sep="")] <-
rowSums(as.matrix(glb_full_DTM_lst[[txt_var]]) != 0)
}
# a working copy of this is reqd in manage.missingdata chunk
union_df <- rbind(mutate(glb_trnent_df, .src = "trnent"),
mutate(glb_newent_df, .src = "newent"))
tmp_entity_df <- glb_entity_df
mrg_id_vars <- ifelse(length(glb_id_vars) > 0, glb_id_vars, ".rownames")
if (mrg_id_vars == ".rownames") {
union_df$.rownames <- rownames(union_df)
tmp_entity_df$.rownames <- rownames(tmp_entity_df)
}
mrg_entity_df <- merge(tmp_entity_df, union_df[, c(".src", mrg_id_vars)])
# Partition again
glb_trnent_df <- subset(mrg_entity_df, .src == "trnent", select=-c(.src))
glb_newent_df <- subset(mrg_entity_df, .src == "newent", select=-c(.src))
if (mrg_id_vars == ".rownames") {
glb_trnent_df <- subset(glb_trnent_df, select=-c(.rownames))
glb_newent_df <- subset(glb_newent_df, select=-c(.rownames))
}
comment(glb_trnent_df) <- "trnent_df"
comment(glb_newent_df) <- "newent_df"
# Generate summaries
# print(summary(glb_entity_df))
# print(sapply(names(glb_entity_df), function(col) sum(is.na(glb_entity_df[, col]))))
# print(summary(glb_trnent_df))
# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(summary(glb_newent_df))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
}## Loading required package: tm
## Loading required package: NLP
##
## Attaching package: 'NLP'
##
## The following object is masked from 'package:ggplot2':
##
## annotate## [1] "Building corpus for title..."
## [1] " Full freqs:"
## <<DocumentTermMatrix (documents: 1860, terms: 2833)>>
## Non-/sparse entries: 23416/5245964
## Sparsity : 100%
## Maximal term length: 49
## Weighting : term frequency (tf)
## [1] " Sparse freqs:"
## <<DocumentTermMatrix (documents: 1860, terms: 31)>>
## Non-/sparse entries: 10684/46976
## Sparsity : 81%
## Maximal term length: 15
## Weighting : term frequency (tf)## Warning: Removed 6 rows containing missing values (geom_path).
## [1] "Building corpus for abstract..."
## [1] " Full freqs:"
## <<DocumentTermMatrix (documents: 1860, terms: 12224)>>
## Non-/sparse entries: 153170/22583470
## Sparsity : 99%
## Maximal term length: 67
## Weighting : term frequency (tf)
## [1] " Sparse freqs:"
## <<DocumentTermMatrix (documents: 1860, terms: 335)>>
## Non-/sparse entries: 92016/531084
## Sparsity : 85%
## Maximal term length: 15
## Weighting : term frequency (tf)## Warning: Removed 6 rows containing missing values (geom_path).
## [1] "Binding DTM for title..."
## [1] "Binding DTM for abstract..."# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
# print(myplot_scatter(glb_trnent_df, "<col1_name>", "<col2_name>", smooth=TRUE))
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all","data.new")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="select_features",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed5 extract.features 3 0 6.239
## elapsed6 select_features 4 0 26.089Step 4: select features
print(glb_feats_df <- myselect_features(entity_df=glb_trnent_df,
exclude_vars_as_features=glb_exclude_vars_as_features,
rsp_var=glb_rsp_var))## Warning in cor(data.matrix(entity_df[, sel_feats]), y =
## as.numeric(entity_df[, : the standard deviation is zero## id cor.y exclude.as.feat
## trial trial 1.000000e+00 1
## T.phase T.phase 4.407918e-01 0
## A.toxic A.toxic 3.546830e-01 0
## A.mgm2 A.mgm2 3.476197e-01 0
## T.metastat T.metastat 3.049464e-01 0
## A.everi A.everi 2.932475e-01 0
## A.median A.median 2.890542e-01 0
## A.rate A.rate 2.658285e-01 0
## T.studi T.studi 2.657631e-01 0
## T.versus T.versus 2.497154e-01 0
## T.advanc T.advanc 2.469661e-01 0
## A.respons A.respons 2.450234e-01 0
## A.metastat A.metastat 2.417899e-01 0
## A.use A.use -2.345315e-01 0
## A.progress A.progress 2.284989e-01 0
## A.advanc A.advanc 2.281995e-01 0
## A.neutropenia A.neutropenia 2.262877e-01 0
## A.partial A.partial 2.185387e-01 0
## A.toler A.toler 2.093203e-01 0
## A.combin A.combin 2.088035e-01 0
## A.firstlin A.firstlin 2.039271e-01 0
## A.regimen A.regimen 2.028112e-01 0
## T.breast T.breast 1.989942e-01 0
## A.cyclophosphamid A.cyclophosphamid 1.958128e-01 0
## T.docetaxel T.docetaxel 1.947844e-01 0
## T.cyclophosphamid T.cyclophosphamid 1.932487e-01 0
## A.overal A.overal 1.927655e-01 0
## A.week A.week 1.918602e-01 0
## A.month A.month 1.887787e-01 0
## A.risk A.risk -1.845209e-01 0
## A.surviv A.surviv 1.844165e-01 0
## T.combin T.combin 1.835246e-01 0
## T.effect T.effect -1.798643e-01 0
## A.day A.day 1.770836e-01 0
## A.predict A.predict -1.756009e-01 0
## A.associ A.associ -1.738833e-01 0
## A.arm A.arm 1.729602e-01 0
## A.docetaxel A.docetaxel 1.721916e-01 0
## T.cancer T.cancer 1.680898e-01 0
## T.iii T.iii 1.663443e-01 0
## T.compar T.compar 1.658208e-01 0
## A.durat A.durat 1.644884e-01 0
## T.plus T.plus 1.614591e-01 0
## A.data A.data -1.605586e-01 0
## A.given A.given 1.577815e-01 0
## A.epirubicin A.epirubicin 1.576660e-01 0
## A.prevent A.prevent -1.556778e-01 0
## A.bone A.bone -1.552270e-01 0
## A.mbc A.mbc 1.540985e-01 0
## A.decreas A.decreas -1.536759e-01 0
## A.previous A.previous 1.535045e-01 0
## A.identifi A.identifi -1.524843e-01 0
## A.baselin A.baselin -1.517685e-01 0
## A.X100 A.X100 1.513446e-01 0
## A.four A.four 1.512749e-01 0
## A.cycl A.cycl 1.511902e-01 0
## A.methotrex A.methotrex 1.508929e-01 0
## T.respons T.respons -1.496979e-01 0
## A.chang A.chang -1.494941e-01 0
## T.group T.group 1.488481e-01 0
## A.measur A.measur -1.483961e-01 0
## A.phase A.phase 1.481848e-01 0
## A.cell A.cell -1.478798e-01 0
## A.reduc A.reduc -1.472907e-01 0
## A.diseas A.diseas 1.454492e-01 0
## A.efficaci A.efficaci 1.446412e-01 0
## A.X500 A.X500 1.433281e-01 0
## A.grade A.grade 1.432705e-01 0
## A.patient A.patient 1.420881e-01 0
## A.marker A.marker -1.391338e-01 0
## A.occur A.occur 1.363943e-01 0
## A.prognost A.prognost -1.363777e-01 0
## A.hundr A.hundr 1.362355e-01 0
## A.examin A.examin -1.361521e-01 0
## A.control A.control -1.360290e-01 0
## A.schedul A.schedul 1.358289e-01 0
## A.particip A.particip -1.330350e-01 0
## A.found A.found -1.328394e-01 0
## T.doxorubicin T.doxorubicin 1.324163e-01 0
## A.express A.express -1.321424e-01 0
## A.score A.score -1.320092e-01 0
## A.either A.either 1.309481e-01 0
## A.high A.high -1.285034e-01 0
## A.dose A.dose 1.284812e-01 0
## A.activ A.activ 1.271691e-01 0
## A.progressionfre A.progressionfre 1.270990e-01 0
## A.trial A.trial -1.258125e-01 0
## A.model A.model -1.234019e-01 0
## A.multivari A.multivari -1.231479e-01 0
## A.placebo A.placebo -1.229875e-01 0
## A.safeti A.safeti 1.217685e-01 0
## A.doxorubicin A.doxorubicin 1.217472e-01 0
## A.level A.level -1.206844e-01 0
## A.seen A.seen 1.203577e-01 0
## A.elig A.elig 1.201664e-01 0
## T.trial T.trial 1.195178e-01 0
## A.increas A.increas -1.192067e-01 0
## A.hematolog A.hematolog 1.187144e-01 0
## A.correl A.correl -1.185756e-01 0
## A.independ A.independ -1.181013e-01 0
## A.plus A.plus 1.174459e-01 0
## A.need A.need -1.172494e-01 0
## A.vomit A.vomit 1.161292e-01 0
## A.respect A.respect 1.158080e-01 0
## A.assess A.assess -1.157562e-01 0
## A.fluorouracil A.fluorouracil 1.156599e-01 0
## A.X5fluorouracil A.X5fluorouracil 1.153122e-01 0
## A.superior A.superior 1.144624e-01 0
## A.inform A.inform -1.140931e-01 0
## A.enter A.enter 1.130515e-01 0
## T.num.chars T.num.chars 1.122127e-01 0
## A.sampl A.sampl -1.119119e-01 0
## A.two A.two 1.118305e-01 0
## A.valu A.valu -1.111873e-01 0
## A.proport A.proport -1.076947e-01 0
## A.object A.object 1.065300e-01 0
## A.analyz A.analyz -1.060879e-01 0
## A.rang A.rang 1.050787e-01 0
## A.factor A.factor -1.049804e-01 0
## A.paclitaxel A.paclitaxel 1.044681e-01 0
## T.num.words T.num.words 1.040679e-01 0
## A.serum A.serum -1.037758e-01 0
## A.three A.three 1.031348e-01 0
## A.tumour A.tumour -1.029714e-01 0
## T.random T.random 1.028374e-01 0
## A.prior A.prior 1.026878e-01 0
## A.cmf A.cmf 1.023838e-01 0
## A.doubleblind A.doubleblind -1.009869e-01 0
## A.can A.can -1.009582e-01 0
## A.qualiti A.qualiti -1.007081e-01 0
## A.tumor A.tumor -9.949416e-02 0
## A.sequenti A.sequenti 9.916268e-02 0
## A.provid A.provid -9.854909e-02 0
## A.oral A.oral 9.728974e-02 0
## A.complet A.complet 9.717265e-02 0
## A.diseasefre A.diseasefre 9.577232e-02 0
## A.achiev A.achiev 9.514020e-02 0
## A.similar A.similar 9.371701e-02 0
## T.randomis T.randomis 9.370110e-02 0
## A.six A.six 9.362652e-02 0
## A.show A.show -9.265780e-02 0
## A.secondari A.secondari 8.944036e-02 0
## A.report A.report -8.926696e-02 0
## A.X005 A.X005 -8.880757e-02 0
## A.analys A.analys -8.879582e-02 0
## A.well A.well 8.870282e-02 0
## A.negat A.negat -8.863047e-02 0
## A.avail A.avail -8.834244e-02 0
## A.analysi A.analysi -8.745562e-02 0
## A.function. A.function. -8.739253e-02 0
## A.type A.type -8.602559e-02 0
## T.num.words.unq T.num.words.unq 8.542797e-02 0
## A.metastas A.metastas -8.500380e-02 0
## A.event A.event 8.489612e-02 0
## A.anthracyclin A.anthracyclin 8.474734e-02 0
## A.death A.death 8.448898e-02 0
## A.test A.test -8.422639e-02 0
## A.longer A.longer 8.421948e-02 0
## A.eight A.eight 8.362672e-02 0
## A.system A.system -8.347057e-02 0
## A.indic A.indic -8.330481e-02 0
## A.may A.may -8.289575e-02 0
## A.wherea A.wherea -8.264291e-02 0
## A.one A.one 8.221967e-02 0
## A.outcom A.outcom -8.083204e-02 0
## A.enrol A.enrol 7.928104e-02 0
## A.effect A.effect -7.906464e-02 0
## A.term A.term 7.818056e-02 0
## A.administ A.administ 7.791871e-02 0
## A.failur A.failur 7.764819e-02 0
## A.size A.size -7.746176e-02 0
## A.life A.life -7.741262e-02 0
## A.common A.common 7.737623e-02 0
## A.relat A.relat -7.729160e-02 0
## A.frequent A.frequent 7.708184e-02 0
## A.also A.also -7.612550e-02 0
## A.clinic A.clinic -7.577756e-02 0
## A.radiotherapi A.radiotherapi -7.542001e-02 0
## A.assign A.assign 7.540280e-02 0
## T.women T.women -7.520353e-02 0
## A.nodeposit A.nodeposit 7.498902e-02 0
## A.mean A.mean -7.485297e-02 0
## A.low A.low -7.478252e-02 0
## A.signific A.signific -7.475443e-02 0
## A.respond A.respond 7.372052e-02 0
## A.first A.first -7.206416e-02 0
## A.select A.select -7.095647e-02 0
## A.endpoint A.endpoint 7.093389e-02 0
## A.stabl A.stabl 7.053656e-02 0
## A.confid A.confid 7.049294e-02 0
## T.earli T.earli -7.039691e-02 0
## A.prolong A.prolong 6.968693e-02 0
## A.lower A.lower -6.958213e-02 0
## A.histolog A.histolog -6.904065e-02 0
## A.continu A.continu 6.889854e-02 0
## A.daili A.daili 6.879414e-02 0
## A.chemotherapi A.chemotherapi -6.848646e-02 0
## A.investig A.investig -6.814008e-02 0
## A.experienc A.experienc 6.719543e-02 0
## A.recurr A.recurr -6.713675e-02 0
## A.follow A.follow 6.707377e-02 0
## A.women A.women -6.706568e-02 0
## A.detect A.detect -6.580512e-02 0
## A.treatment A.treatment 6.578098e-02 0
## A.end A.end 6.542423e-02 0
## A.import A.import -6.500431e-02 0
## A.potenti A.potenti -6.439343e-02 0
## A.X0001 A.X0001 -6.383307e-02 0
## A.design A.design -6.374264e-02 0
## A.remain A.remain -6.292028e-02 0
## A.time A.time 6.178757e-02 0
## A.receiv A.receiv 6.090943e-02 0
## A.dfs A.dfs 6.076035e-02 0
## A.primari A.primari 6.020830e-02 0
## A.within A.within -5.934658e-02 0
## A.interv A.interv 5.928858e-02 0
## A.compar A.compar 5.923358e-02 0
## A.total A.total 5.916846e-02 0
## T.treatment T.treatment 5.769991e-02 0
## A.suggest A.suggest -5.757780e-02 0
## A.obtain A.obtain -5.756790e-02 0
## A.find A.find -5.637269e-02 0
## A.case A.case -5.636252e-02 0
## A.adjuv A.adjuv -5.608234e-02 0
## A.tissu A.tissu -5.595253e-02 0
## A.cancer A.cancer -5.581714e-02 0
## A.intraven A.intraven 5.564821e-02 0
## T.tamoxifen T.tamoxifen 5.564564e-02 0
## T.result T.result 5.427221e-02 0
## A.cours A.cours 5.392045e-02 0
## A.estrogen A.estrogen -5.357349e-02 0
## A.observ A.observ 5.273205e-02 0
## A.breast A.breast -5.233490e-02 0
## A.second A.second -5.198940e-02 0
## A.advers A.advers 5.167782e-02 0
## T.clinic T.clinic -5.148394e-02 0
## A.evalu A.evalu 5.098452e-02 0
## A.reduct A.reduct -5.075505e-02 0
## A.period A.period -5.074228e-02 0
## A.hormon A.hormon -5.048861e-02 0
## A.versus A.versus 5.026899e-02 0
## A.shown A.shown -4.986612e-02 0
## A.pretreat A.pretreat 4.965851e-02 0
## A.point A.point 4.946137e-02 0
## A.larg A.larg -4.945536e-02 0
## A.perform A.perform -4.866672e-02 0
## T.postmenopaus T.postmenopaus 4.801689e-02 0
## A.possibl A.possibl -4.794828e-02 0
## A.aim A.aim -4.785872e-02 0
## A.present A.present -4.695298e-02 0
## A.includ A.includ 4.584644e-02 0
## A.seven A.seven 4.541994e-02 0
## A.nausea A.nausea 4.537254e-02 0
## A.main A.main -4.489330e-02 0
## .rnorm .rnorm 4.466651e-02 0
## A.five A.five 4.460546e-02 0
## A.local A.local -4.366301e-02 0
## A.aromatas A.aromatas -4.322202e-02 0
## A.progesteron A.progesteron -4.318635e-02 0
## A.posit A.posit -4.317584e-02 0
## A.determin A.determin -4.239248e-02 0
## A.incid A.incid -4.227791e-02 0
## A.start A.start -4.140127e-02 0
## A.better A.better 4.133124e-02 0
## A.status A.status -4.120076e-02 0
## A.her2 A.her2 -4.085805e-02 0
## A.multicent A.multicent 4.053909e-02 0
## A.set A.set -3.997140e-02 0
## A.oper A.oper 3.993046e-02 0
## A.stage A.stage 3.975581e-02 0
## A.alon A.alon 3.907515e-02 0
## A.patholog A.patholog -3.751957e-02 0
## A.less A.less 3.732604e-02 0
## A.random A.random 3.663241e-02 0
## A.although A.although 3.604544e-02 0
## A.method A.method -3.582004e-02 0
## A.addit A.addit 3.510086e-02 0
## A.carcinoma A.carcinoma -3.493790e-02 0
## A.has.http A.has.http -3.472002e-02 0
## A.base A.base -3.357553e-02 0
## A.X001 A.X001 -3.311267e-02 0
## A.earli A.earli -3.276589e-02 0
## A.agent A.agent 3.257901e-02 0
## A.general A.general -3.234203e-02 0
## A.age A.age -3.188645e-02 0
## A.per A.per -3.173574e-02 0
## A.due A.due 3.164831e-02 0
## A.regress A.regress -3.139311e-02 0
## A.prospect A.prospect -3.102924e-02 0
## A.andor A.andor 3.086932e-02 0
## A.comparison A.comparison -3.040785e-02 0
## A.lymph A.lymph 3.005185e-02 0
## A.develop A.develop -3.000773e-02 0
## A.followup A.followup 2.973279e-02 0
## A.number A.number -2.963870e-02 0
## T.adjuv T.adjuv -2.813822e-02 0
## A.support A.support -2.793158e-02 0
## A.postmenopaus A.postmenopaus 2.765137e-02 0
## A.involv A.involv 2.700336e-02 0
## A.side A.side 2.689790e-02 0
## A.howev A.howev -2.671381e-02 0
## A.standard A.standard 2.641969e-02 0
## A.studi A.studi -2.594717e-02 0
## A.therapi A.therapi -2.549221e-02 0
## A.receptorposit A.receptorposit 2.492618e-02 0
## A.conduct A.conduct 2.467401e-02 0
## A.greater A.greater -2.407688e-02 0
## A.receptor A.receptor -2.389969e-02 0
## A.subgroup A.subgroup -2.338162e-02 0
## A.trend A.trend -2.337564e-02 0
## A.defin A.defin -2.332432e-02 0
## A.requir A.requir -2.281593e-02 0
## A.regard A.regard -2.212204e-02 0
## A.neoadjuv A.neoadjuv 2.210203e-02 0
## A.major A.major 2.209530e-02 0
## A.distant A.distant -2.186214e-02 0
## A.postop A.postop -2.083001e-02 0
## A.administr A.administr 2.040770e-02 0
## A.new A.new -2.033234e-02 0
## A.estim A.estim -1.960142e-02 0
## A.endocrin A.endocrin -1.936997e-02 0
## A.caus A.caus -1.915695e-02 0
## A.limit A.limit 1.884501e-02 0
## A.consist A.consist 1.855504e-02 0
## A.num.words A.num.words -1.799787e-02 0
## A.differ A.differ 1.797677e-02 0
## A.num.words.unq A.num.words.unq 1.784364e-02 0
## A.singl A.singl 1.695022e-02 0
## A.hazard A.hazard 1.634076e-02 0
## A.axillari A.axillari -1.556611e-02 0
## A.sever A.sever 1.516401e-02 0
## A.node A.node -1.504773e-02 0
## A.result A.result 1.442596e-02 0
## A.randomis A.randomis 1.425652e-02 0
## A.consid A.consid -1.417986e-02 0
## A.least A.least -1.387475e-02 0
## A.mastectomi A.mastectomi -1.359074e-02 0
## A.group A.group -1.337887e-02 0
## A.initi A.initi -1.332423e-02 0
## A.appear A.appear 1.314392e-02 0
## A.popul A.popul -1.296710e-02 0
## A.whether A.whether -1.291745e-02 0
## A.characterist A.characterist -1.258143e-02 0
## A.inhibitor A.inhibitor -1.239898e-02 0
## A.confirm A.confirm -1.221603e-02 0
## A.treat A.treat 1.103432e-02 0
## A.profil A.profil 1.092909e-02 0
## T.chemotherapi T.chemotherapi -9.909758e-03 0
## A.growth A.growth -9.887605e-03 0
## A.year A.year 9.566746e-03 0
## A.benefit A.benefit -9.524764e-03 0
## A.background A.background -9.175455e-03 0
## A.accord A.accord 9.173456e-03 0
## T.patient T.patient 7.667604e-03 0
## A.site A.site 7.553580e-03 0
## A.num.chars A.num.chars -7.349760e-03 0
## A.improv A.improv 7.017871e-03 0
## A.tamoxifen A.tamoxifen 7.004532e-03 0
## A.ratio A.ratio 6.872134e-03 0
## A.premenopaus A.premenopaus -6.600118e-03 0
## A.purpos A.purpos -6.501960e-03 0
## A.evid A.evid 5.914321e-03 0
## A.higher A.higher -3.196262e-03 0
## T.therapi T.therapi -3.064253e-03 0
## A.iii A.iii 3.062659e-03 0
## A.infus A.infus -2.691231e-03 0
## A.conclus A.conclus 2.541412e-03 0
## A.drug A.drug -2.504858e-03 0
## A.surgeri A.surgeri -2.256826e-03 0
## A.without A.without -1.584596e-03 0
## A.demonstr A.demonstr 1.364004e-03 0
## A.statist A.statist 1.141908e-03 0
## A.human A.human -9.121557e-04 0
## A.among A.among 3.632848e-04 0
## A.relaps A.relaps -7.929870e-06 0
## T.has.http T.has.http NA 0
## cor.y.abs
## trial 1.000000e+00
## T.phase 4.407918e-01
## A.toxic 3.546830e-01
## A.mgm2 3.476197e-01
## T.metastat 3.049464e-01
## A.everi 2.932475e-01
## A.median 2.890542e-01
## A.rate 2.658285e-01
## T.studi 2.657631e-01
## T.versus 2.497154e-01
## T.advanc 2.469661e-01
## A.respons 2.450234e-01
## A.metastat 2.417899e-01
## A.use 2.345315e-01
## A.progress 2.284989e-01
## A.advanc 2.281995e-01
## A.neutropenia 2.262877e-01
## A.partial 2.185387e-01
## A.toler 2.093203e-01
## A.combin 2.088035e-01
## A.firstlin 2.039271e-01
## A.regimen 2.028112e-01
## T.breast 1.989942e-01
## A.cyclophosphamid 1.958128e-01
## T.docetaxel 1.947844e-01
## T.cyclophosphamid 1.932487e-01
## A.overal 1.927655e-01
## A.week 1.918602e-01
## A.month 1.887787e-01
## A.risk 1.845209e-01
## A.surviv 1.844165e-01
## T.combin 1.835246e-01
## T.effect 1.798643e-01
## A.day 1.770836e-01
## A.predict 1.756009e-01
## A.associ 1.738833e-01
## A.arm 1.729602e-01
## A.docetaxel 1.721916e-01
## T.cancer 1.680898e-01
## T.iii 1.663443e-01
## T.compar 1.658208e-01
## A.durat 1.644884e-01
## T.plus 1.614591e-01
## A.data 1.605586e-01
## A.given 1.577815e-01
## A.epirubicin 1.576660e-01
## A.prevent 1.556778e-01
## A.bone 1.552270e-01
## A.mbc 1.540985e-01
## A.decreas 1.536759e-01
## A.previous 1.535045e-01
## A.identifi 1.524843e-01
## A.baselin 1.517685e-01
## A.X100 1.513446e-01
## A.four 1.512749e-01
## A.cycl 1.511902e-01
## A.methotrex 1.508929e-01
## T.respons 1.496979e-01
## A.chang 1.494941e-01
## T.group 1.488481e-01
## A.measur 1.483961e-01
## A.phase 1.481848e-01
## A.cell 1.478798e-01
## A.reduc 1.472907e-01
## A.diseas 1.454492e-01
## A.efficaci 1.446412e-01
## A.X500 1.433281e-01
## A.grade 1.432705e-01
## A.patient 1.420881e-01
## A.marker 1.391338e-01
## A.occur 1.363943e-01
## A.prognost 1.363777e-01
## A.hundr 1.362355e-01
## A.examin 1.361521e-01
## A.control 1.360290e-01
## A.schedul 1.358289e-01
## A.particip 1.330350e-01
## A.found 1.328394e-01
## T.doxorubicin 1.324163e-01
## A.express 1.321424e-01
## A.score 1.320092e-01
## A.either 1.309481e-01
## A.high 1.285034e-01
## A.dose 1.284812e-01
## A.activ 1.271691e-01
## A.progressionfre 1.270990e-01
## A.trial 1.258125e-01
## A.model 1.234019e-01
## A.multivari 1.231479e-01
## A.placebo 1.229875e-01
## A.safeti 1.217685e-01
## A.doxorubicin 1.217472e-01
## A.level 1.206844e-01
## A.seen 1.203577e-01
## A.elig 1.201664e-01
## T.trial 1.195178e-01
## A.increas 1.192067e-01
## A.hematolog 1.187144e-01
## A.correl 1.185756e-01
## A.independ 1.181013e-01
## A.plus 1.174459e-01
## A.need 1.172494e-01
## A.vomit 1.161292e-01
## A.respect 1.158080e-01
## A.assess 1.157562e-01
## A.fluorouracil 1.156599e-01
## A.X5fluorouracil 1.153122e-01
## A.superior 1.144624e-01
## A.inform 1.140931e-01
## A.enter 1.130515e-01
## T.num.chars 1.122127e-01
## A.sampl 1.119119e-01
## A.two 1.118305e-01
## A.valu 1.111873e-01
## A.proport 1.076947e-01
## A.object 1.065300e-01
## A.analyz 1.060879e-01
## A.rang 1.050787e-01
## A.factor 1.049804e-01
## A.paclitaxel 1.044681e-01
## T.num.words 1.040679e-01
## A.serum 1.037758e-01
## A.three 1.031348e-01
## A.tumour 1.029714e-01
## T.random 1.028374e-01
## A.prior 1.026878e-01
## A.cmf 1.023838e-01
## A.doubleblind 1.009869e-01
## A.can 1.009582e-01
## A.qualiti 1.007081e-01
## A.tumor 9.949416e-02
## A.sequenti 9.916268e-02
## A.provid 9.854909e-02
## A.oral 9.728974e-02
## A.complet 9.717265e-02
## A.diseasefre 9.577232e-02
## A.achiev 9.514020e-02
## A.similar 9.371701e-02
## T.randomis 9.370110e-02
## A.six 9.362652e-02
## A.show 9.265780e-02
## A.secondari 8.944036e-02
## A.report 8.926696e-02
## A.X005 8.880757e-02
## A.analys 8.879582e-02
## A.well 8.870282e-02
## A.negat 8.863047e-02
## A.avail 8.834244e-02
## A.analysi 8.745562e-02
## A.function. 8.739253e-02
## A.type 8.602559e-02
## T.num.words.unq 8.542797e-02
## A.metastas 8.500380e-02
## A.event 8.489612e-02
## A.anthracyclin 8.474734e-02
## A.death 8.448898e-02
## A.test 8.422639e-02
## A.longer 8.421948e-02
## A.eight 8.362672e-02
## A.system 8.347057e-02
## A.indic 8.330481e-02
## A.may 8.289575e-02
## A.wherea 8.264291e-02
## A.one 8.221967e-02
## A.outcom 8.083204e-02
## A.enrol 7.928104e-02
## A.effect 7.906464e-02
## A.term 7.818056e-02
## A.administ 7.791871e-02
## A.failur 7.764819e-02
## A.size 7.746176e-02
## A.life 7.741262e-02
## A.common 7.737623e-02
## A.relat 7.729160e-02
## A.frequent 7.708184e-02
## A.also 7.612550e-02
## A.clinic 7.577756e-02
## A.radiotherapi 7.542001e-02
## A.assign 7.540280e-02
## T.women 7.520353e-02
## A.nodeposit 7.498902e-02
## A.mean 7.485297e-02
## A.low 7.478252e-02
## A.signific 7.475443e-02
## A.respond 7.372052e-02
## A.first 7.206416e-02
## A.select 7.095647e-02
## A.endpoint 7.093389e-02
## A.stabl 7.053656e-02
## A.confid 7.049294e-02
## T.earli 7.039691e-02
## A.prolong 6.968693e-02
## A.lower 6.958213e-02
## A.histolog 6.904065e-02
## A.continu 6.889854e-02
## A.daili 6.879414e-02
## A.chemotherapi 6.848646e-02
## A.investig 6.814008e-02
## A.experienc 6.719543e-02
## A.recurr 6.713675e-02
## A.follow 6.707377e-02
## A.women 6.706568e-02
## A.detect 6.580512e-02
## A.treatment 6.578098e-02
## A.end 6.542423e-02
## A.import 6.500431e-02
## A.potenti 6.439343e-02
## A.X0001 6.383307e-02
## A.design 6.374264e-02
## A.remain 6.292028e-02
## A.time 6.178757e-02
## A.receiv 6.090943e-02
## A.dfs 6.076035e-02
## A.primari 6.020830e-02
## A.within 5.934658e-02
## A.interv 5.928858e-02
## A.compar 5.923358e-02
## A.total 5.916846e-02
## T.treatment 5.769991e-02
## A.suggest 5.757780e-02
## A.obtain 5.756790e-02
## A.find 5.637269e-02
## A.case 5.636252e-02
## A.adjuv 5.608234e-02
## A.tissu 5.595253e-02
## A.cancer 5.581714e-02
## A.intraven 5.564821e-02
## T.tamoxifen 5.564564e-02
## T.result 5.427221e-02
## A.cours 5.392045e-02
## A.estrogen 5.357349e-02
## A.observ 5.273205e-02
## A.breast 5.233490e-02
## A.second 5.198940e-02
## A.advers 5.167782e-02
## T.clinic 5.148394e-02
## A.evalu 5.098452e-02
## A.reduct 5.075505e-02
## A.period 5.074228e-02
## A.hormon 5.048861e-02
## A.versus 5.026899e-02
## A.shown 4.986612e-02
## A.pretreat 4.965851e-02
## A.point 4.946137e-02
## A.larg 4.945536e-02
## A.perform 4.866672e-02
## T.postmenopaus 4.801689e-02
## A.possibl 4.794828e-02
## A.aim 4.785872e-02
## A.present 4.695298e-02
## A.includ 4.584644e-02
## A.seven 4.541994e-02
## A.nausea 4.537254e-02
## A.main 4.489330e-02
## .rnorm 4.466651e-02
## A.five 4.460546e-02
## A.local 4.366301e-02
## A.aromatas 4.322202e-02
## A.progesteron 4.318635e-02
## A.posit 4.317584e-02
## A.determin 4.239248e-02
## A.incid 4.227791e-02
## A.start 4.140127e-02
## A.better 4.133124e-02
## A.status 4.120076e-02
## A.her2 4.085805e-02
## A.multicent 4.053909e-02
## A.set 3.997140e-02
## A.oper 3.993046e-02
## A.stage 3.975581e-02
## A.alon 3.907515e-02
## A.patholog 3.751957e-02
## A.less 3.732604e-02
## A.random 3.663241e-02
## A.although 3.604544e-02
## A.method 3.582004e-02
## A.addit 3.510086e-02
## A.carcinoma 3.493790e-02
## A.has.http 3.472002e-02
## A.base 3.357553e-02
## A.X001 3.311267e-02
## A.earli 3.276589e-02
## A.agent 3.257901e-02
## A.general 3.234203e-02
## A.age 3.188645e-02
## A.per 3.173574e-02
## A.due 3.164831e-02
## A.regress 3.139311e-02
## A.prospect 3.102924e-02
## A.andor 3.086932e-02
## A.comparison 3.040785e-02
## A.lymph 3.005185e-02
## A.develop 3.000773e-02
## A.followup 2.973279e-02
## A.number 2.963870e-02
## T.adjuv 2.813822e-02
## A.support 2.793158e-02
## A.postmenopaus 2.765137e-02
## A.involv 2.700336e-02
## A.side 2.689790e-02
## A.howev 2.671381e-02
## A.standard 2.641969e-02
## A.studi 2.594717e-02
## A.therapi 2.549221e-02
## A.receptorposit 2.492618e-02
## A.conduct 2.467401e-02
## A.greater 2.407688e-02
## A.receptor 2.389969e-02
## A.subgroup 2.338162e-02
## A.trend 2.337564e-02
## A.defin 2.332432e-02
## A.requir 2.281593e-02
## A.regard 2.212204e-02
## A.neoadjuv 2.210203e-02
## A.major 2.209530e-02
## A.distant 2.186214e-02
## A.postop 2.083001e-02
## A.administr 2.040770e-02
## A.new 2.033234e-02
## A.estim 1.960142e-02
## A.endocrin 1.936997e-02
## A.caus 1.915695e-02
## A.limit 1.884501e-02
## A.consist 1.855504e-02
## A.num.words 1.799787e-02
## A.differ 1.797677e-02
## A.num.words.unq 1.784364e-02
## A.singl 1.695022e-02
## A.hazard 1.634076e-02
## A.axillari 1.556611e-02
## A.sever 1.516401e-02
## A.node 1.504773e-02
## A.result 1.442596e-02
## A.randomis 1.425652e-02
## A.consid 1.417986e-02
## A.least 1.387475e-02
## A.mastectomi 1.359074e-02
## A.group 1.337887e-02
## A.initi 1.332423e-02
## A.appear 1.314392e-02
## A.popul 1.296710e-02
## A.whether 1.291745e-02
## A.characterist 1.258143e-02
## A.inhibitor 1.239898e-02
## A.confirm 1.221603e-02
## A.treat 1.103432e-02
## A.profil 1.092909e-02
## T.chemotherapi 9.909758e-03
## A.growth 9.887605e-03
## A.year 9.566746e-03
## A.benefit 9.524764e-03
## A.background 9.175455e-03
## A.accord 9.173456e-03
## T.patient 7.667604e-03
## A.site 7.553580e-03
## A.num.chars 7.349760e-03
## A.improv 7.017871e-03
## A.tamoxifen 7.004532e-03
## A.ratio 6.872134e-03
## A.premenopaus 6.600118e-03
## A.purpos 6.501960e-03
## A.evid 5.914321e-03
## A.higher 3.196262e-03
## T.therapi 3.064253e-03
## A.iii 3.062659e-03
## A.infus 2.691231e-03
## A.conclus 2.541412e-03
## A.drug 2.504858e-03
## A.surgeri 2.256826e-03
## A.without 1.584596e-03
## A.demonstr 1.364004e-03
## A.statist 1.141908e-03
## A.human 9.121557e-04
## A.among 3.632848e-04
## A.relaps 7.929870e-06
## T.has.http NAglb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="remove_correlated_features",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor
## elapsed6 select_features 4 0
## elapsed7 remove_correlated_features 4 1
## elapsed
## elapsed6 26.089
## elapsed7 27.537Step 4.1: remove correlated features
print(glb_feats_df <- orderBy(~-cor.y,
myfind_cor_features(feats_df=glb_feats_df, entity_df=glb_trnent_df,
rsp_var=glb_rsp_var,
checkConditionalX=(glb_is_classification && glb_is_binomial))))## Loading required package: caret
## Loading required package: lattice
##
## Attaching package: 'caret'
##
## The following object is masked from 'package:survival':
##
## cluster## [1] "cor(T.num.words, T.num.words.unq)=0.9838"
## [1] "cor(trial.fctr, T.num.words)=0.1041"
## [1] "cor(trial.fctr, T.num.words.unq)=0.0854"## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified T.num.words.unq as highly correlated with
## T.num.words
## [1] "cor(T.num.chars, T.num.words)=0.9543"
## [1] "cor(trial.fctr, T.num.chars)=0.1122"
## [1] "cor(trial.fctr, T.num.words)=0.1041"## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified T.num.words as highly correlated with
## T.num.chars
## [1] "cor(A.life, A.qualiti)=0.8704"
## [1] "cor(trial.fctr, A.life)=-0.0774"
## [1] "cor(trial.fctr, A.qualiti)=-0.1007"## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.life as highly correlated with A.qualiti
## [1] "cor(A.breast, A.cancer)=0.7698"
## [1] "cor(trial.fctr, A.breast)=-0.0523"
## [1] "cor(trial.fctr, A.cancer)=-0.0558"## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.breast as highly correlated with A.cancer
## [1] "cor(T.breast, T.cancer)=0.7283"
## [1] "cor(trial.fctr, T.breast)=0.1990"
## [1] "cor(trial.fctr, T.cancer)=0.1681"## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified T.cancer as highly correlated with T.breast
## [1] "cor(A.nausea, A.vomit)=0.7231"
## [1] "cor(trial.fctr, A.nausea)=0.0454"
## [1] "cor(trial.fctr, A.vomit)=0.1161"## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.nausea as highly correlated with A.vomit
## [1] "cor(A.end, A.point)=0.7067"
## [1] "cor(trial.fctr, A.end)=0.0654"
## [1] "cor(trial.fctr, A.point)=0.0495"## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?
## geom_smooth: Only one unique x value each group.Maybe you want aes(group = 1)?## Warning in myfind_cor_features(feats_df = glb_feats_df, entity_df =
## glb_trnent_df, : Identified A.point as highly correlated with A.end
## id cor.y exclude.as.feat
## trial trial 1.000000e+00 1
## T.phase T.phase 4.407918e-01 0
## A.toxic A.toxic 3.546830e-01 0
## A.mgm2 A.mgm2 3.476197e-01 0
## T.metastat T.metastat 3.049464e-01 0
## A.everi A.everi 2.932475e-01 0
## A.median A.median 2.890542e-01 0
## A.rate A.rate 2.658285e-01 0
## T.studi T.studi 2.657631e-01 0
## T.versus T.versus 2.497154e-01 0
## T.advanc T.advanc 2.469661e-01 0
## A.respons A.respons 2.450234e-01 0
## A.metastat A.metastat 2.417899e-01 0
## A.progress A.progress 2.284989e-01 0
## A.advanc A.advanc 2.281995e-01 0
## A.neutropenia A.neutropenia 2.262877e-01 0
## A.partial A.partial 2.185387e-01 0
## A.toler A.toler 2.093203e-01 0
## A.combin A.combin 2.088035e-01 0
## A.firstlin A.firstlin 2.039271e-01 0
## A.regimen A.regimen 2.028112e-01 0
## T.breast T.breast 1.989942e-01 0
## A.cyclophosphamid A.cyclophosphamid 1.958128e-01 0
## T.docetaxel T.docetaxel 1.947844e-01 0
## T.cyclophosphamid T.cyclophosphamid 1.932487e-01 0
## A.overal A.overal 1.927655e-01 0
## A.week A.week 1.918602e-01 0
## A.month A.month 1.887787e-01 0
## A.surviv A.surviv 1.844165e-01 0
## T.combin T.combin 1.835246e-01 0
## A.day A.day 1.770836e-01 0
## A.arm A.arm 1.729602e-01 0
## A.docetaxel A.docetaxel 1.721916e-01 0
## T.cancer T.cancer 1.680898e-01 0
## T.iii T.iii 1.663443e-01 0
## T.compar T.compar 1.658208e-01 0
## A.durat A.durat 1.644884e-01 0
## T.plus T.plus 1.614591e-01 0
## A.given A.given 1.577815e-01 0
## A.epirubicin A.epirubicin 1.576660e-01 0
## A.mbc A.mbc 1.540985e-01 0
## A.previous A.previous 1.535045e-01 0
## A.X100 A.X100 1.513446e-01 0
## A.four A.four 1.512749e-01 0
## A.cycl A.cycl 1.511902e-01 0
## A.methotrex A.methotrex 1.508929e-01 0
## T.group T.group 1.488481e-01 0
## A.phase A.phase 1.481848e-01 0
## A.diseas A.diseas 1.454492e-01 0
## A.efficaci A.efficaci 1.446412e-01 0
## A.X500 A.X500 1.433281e-01 0
## A.grade A.grade 1.432705e-01 0
## A.patient A.patient 1.420881e-01 0
## A.occur A.occur 1.363943e-01 0
## A.hundr A.hundr 1.362355e-01 0
## A.schedul A.schedul 1.358289e-01 0
## T.doxorubicin T.doxorubicin 1.324163e-01 0
## A.either A.either 1.309481e-01 0
## A.dose A.dose 1.284812e-01 0
## A.activ A.activ 1.271691e-01 0
## A.progressionfre A.progressionfre 1.270990e-01 0
## A.safeti A.safeti 1.217685e-01 0
## A.doxorubicin A.doxorubicin 1.217472e-01 0
## A.seen A.seen 1.203577e-01 0
## A.elig A.elig 1.201664e-01 0
## T.trial T.trial 1.195178e-01 0
## A.hematolog A.hematolog 1.187144e-01 0
## A.plus A.plus 1.174459e-01 0
## A.vomit A.vomit 1.161292e-01 0
## A.respect A.respect 1.158080e-01 0
## A.fluorouracil A.fluorouracil 1.156599e-01 0
## A.X5fluorouracil A.X5fluorouracil 1.153122e-01 0
## A.superior A.superior 1.144624e-01 0
## A.enter A.enter 1.130515e-01 0
## T.num.chars T.num.chars 1.122127e-01 0
## A.two A.two 1.118305e-01 0
## A.object A.object 1.065300e-01 0
## A.rang A.rang 1.050787e-01 0
## A.paclitaxel A.paclitaxel 1.044681e-01 0
## T.num.words T.num.words 1.040679e-01 0
## A.three A.three 1.031348e-01 0
## T.random T.random 1.028374e-01 0
## A.prior A.prior 1.026878e-01 0
## A.cmf A.cmf 1.023838e-01 0
## A.sequenti A.sequenti 9.916268e-02 0
## A.oral A.oral 9.728974e-02 0
## A.complet A.complet 9.717265e-02 0
## A.diseasefre A.diseasefre 9.577232e-02 0
## A.achiev A.achiev 9.514020e-02 0
## A.similar A.similar 9.371701e-02 0
## T.randomis T.randomis 9.370110e-02 0
## A.six A.six 9.362652e-02 0
## A.secondari A.secondari 8.944036e-02 0
## A.well A.well 8.870282e-02 0
## T.num.words.unq T.num.words.unq 8.542797e-02 0
## A.event A.event 8.489612e-02 0
## A.anthracyclin A.anthracyclin 8.474734e-02 0
## A.death A.death 8.448898e-02 0
## A.longer A.longer 8.421948e-02 0
## A.eight A.eight 8.362672e-02 0
## A.one A.one 8.221967e-02 0
## A.enrol A.enrol 7.928104e-02 0
## A.term A.term 7.818056e-02 0
## A.administ A.administ 7.791871e-02 0
## A.failur A.failur 7.764819e-02 0
## A.common A.common 7.737623e-02 0
## A.frequent A.frequent 7.708184e-02 0
## A.assign A.assign 7.540280e-02 0
## A.nodeposit A.nodeposit 7.498902e-02 0
## A.respond A.respond 7.372052e-02 0
## A.endpoint A.endpoint 7.093389e-02 0
## A.stabl A.stabl 7.053656e-02 0
## A.confid A.confid 7.049294e-02 0
## A.prolong A.prolong 6.968693e-02 0
## A.continu A.continu 6.889854e-02 0
## A.daili A.daili 6.879414e-02 0
## A.experienc A.experienc 6.719543e-02 0
## A.follow A.follow 6.707377e-02 0
## A.treatment A.treatment 6.578098e-02 0
## A.end A.end 6.542423e-02 0
## A.time A.time 6.178757e-02 0
## A.receiv A.receiv 6.090943e-02 0
## A.dfs A.dfs 6.076035e-02 0
## A.primari A.primari 6.020830e-02 0
## A.interv A.interv 5.928858e-02 0
## A.compar A.compar 5.923358e-02 0
## A.total A.total 5.916846e-02 0
## T.treatment T.treatment 5.769991e-02 0
## A.intraven A.intraven 5.564821e-02 0
## T.tamoxifen T.tamoxifen 5.564564e-02 0
## T.result T.result 5.427221e-02 0
## A.cours A.cours 5.392045e-02 0
## A.observ A.observ 5.273205e-02 0
## A.advers A.advers 5.167782e-02 0
## A.evalu A.evalu 5.098452e-02 0
## A.versus A.versus 5.026899e-02 0
## A.pretreat A.pretreat 4.965851e-02 0
## A.point A.point 4.946137e-02 0
## T.postmenopaus T.postmenopaus 4.801689e-02 0
## A.includ A.includ 4.584644e-02 0
## A.seven A.seven 4.541994e-02 0
## A.nausea A.nausea 4.537254e-02 0
## .rnorm .rnorm 4.466651e-02 0
## A.five A.five 4.460546e-02 0
## A.better A.better 4.133124e-02 0
## A.multicent A.multicent 4.053909e-02 0
## A.oper A.oper 3.993046e-02 0
## A.stage A.stage 3.975581e-02 0
## A.alon A.alon 3.907515e-02 0
## A.less A.less 3.732604e-02 0
## A.random A.random 3.663241e-02 0
## A.although A.although 3.604544e-02 0
## A.addit A.addit 3.510086e-02 0
## A.agent A.agent 3.257901e-02 0
## A.due A.due 3.164831e-02 0
## A.andor A.andor 3.086932e-02 0
## A.lymph A.lymph 3.005185e-02 0
## A.followup A.followup 2.973279e-02 0
## A.postmenopaus A.postmenopaus 2.765137e-02 0
## A.involv A.involv 2.700336e-02 0
## A.side A.side 2.689790e-02 0
## A.standard A.standard 2.641969e-02 0
## A.receptorposit A.receptorposit 2.492618e-02 0
## A.conduct A.conduct 2.467401e-02 0
## A.neoadjuv A.neoadjuv 2.210203e-02 0
## A.major A.major 2.209530e-02 0
## A.administr A.administr 2.040770e-02 0
## A.limit A.limit 1.884501e-02 0
## A.consist A.consist 1.855504e-02 0
## A.differ A.differ 1.797677e-02 0
## A.num.words.unq A.num.words.unq 1.784364e-02 0
## A.singl A.singl 1.695022e-02 0
## A.hazard A.hazard 1.634076e-02 0
## A.sever A.sever 1.516401e-02 0
## A.result A.result 1.442596e-02 0
## A.randomis A.randomis 1.425652e-02 0
## A.appear A.appear 1.314392e-02 0
## A.treat A.treat 1.103432e-02 0
## A.profil A.profil 1.092909e-02 0
## A.year A.year 9.566746e-03 0
## A.accord A.accord 9.173456e-03 0
## T.patient T.patient 7.667604e-03 0
## A.site A.site 7.553580e-03 0
## A.improv A.improv 7.017871e-03 0
## A.tamoxifen A.tamoxifen 7.004532e-03 0
## A.ratio A.ratio 6.872134e-03 0
## A.evid A.evid 5.914321e-03 0
## A.iii A.iii 3.062659e-03 0
## A.conclus A.conclus 2.541412e-03 0
## A.demonstr A.demonstr 1.364004e-03 0
## A.statist A.statist 1.141908e-03 0
## A.among A.among 3.632848e-04 0
## A.relaps A.relaps -7.929870e-06 0
## A.human A.human -9.121557e-04 0
## A.without A.without -1.584596e-03 0
## A.surgeri A.surgeri -2.256826e-03 0
## A.drug A.drug -2.504858e-03 0
## A.infus A.infus -2.691231e-03 0
## T.therapi T.therapi -3.064253e-03 0
## A.higher A.higher -3.196262e-03 0
## A.purpos A.purpos -6.501960e-03 0
## A.premenopaus A.premenopaus -6.600118e-03 0
## A.num.chars A.num.chars -7.349760e-03 0
## A.background A.background -9.175455e-03 0
## A.benefit A.benefit -9.524764e-03 0
## A.growth A.growth -9.887605e-03 0
## T.chemotherapi T.chemotherapi -9.909758e-03 0
## A.confirm A.confirm -1.221603e-02 0
## A.inhibitor A.inhibitor -1.239898e-02 0
## A.characterist A.characterist -1.258143e-02 0
## A.whether A.whether -1.291745e-02 0
## A.popul A.popul -1.296710e-02 0
## A.initi A.initi -1.332423e-02 0
## A.group A.group -1.337887e-02 0
## A.mastectomi A.mastectomi -1.359074e-02 0
## A.least A.least -1.387475e-02 0
## A.consid A.consid -1.417986e-02 0
## A.node A.node -1.504773e-02 0
## A.axillari A.axillari -1.556611e-02 0
## A.num.words A.num.words -1.799787e-02 0
## A.caus A.caus -1.915695e-02 0
## A.endocrin A.endocrin -1.936997e-02 0
## A.estim A.estim -1.960142e-02 0
## A.new A.new -2.033234e-02 0
## A.postop A.postop -2.083001e-02 0
## A.distant A.distant -2.186214e-02 0
## A.regard A.regard -2.212204e-02 0
## A.requir A.requir -2.281593e-02 0
## A.defin A.defin -2.332432e-02 0
## A.trend A.trend -2.337564e-02 0
## A.subgroup A.subgroup -2.338162e-02 0
## A.receptor A.receptor -2.389969e-02 0
## A.greater A.greater -2.407688e-02 0
## A.therapi A.therapi -2.549221e-02 0
## A.studi A.studi -2.594717e-02 0
## A.howev A.howev -2.671381e-02 0
## A.support A.support -2.793158e-02 0
## T.adjuv T.adjuv -2.813822e-02 0
## A.number A.number -2.963870e-02 0
## A.develop A.develop -3.000773e-02 0
## A.comparison A.comparison -3.040785e-02 0
## A.prospect A.prospect -3.102924e-02 0
## A.regress A.regress -3.139311e-02 0
## A.per A.per -3.173574e-02 0
## A.age A.age -3.188645e-02 0
## A.general A.general -3.234203e-02 0
## A.earli A.earli -3.276589e-02 0
## A.X001 A.X001 -3.311267e-02 0
## A.base A.base -3.357553e-02 0
## A.has.http A.has.http -3.472002e-02 0
## A.carcinoma A.carcinoma -3.493790e-02 0
## A.method A.method -3.582004e-02 0
## A.patholog A.patholog -3.751957e-02 0
## A.set A.set -3.997140e-02 0
## A.her2 A.her2 -4.085805e-02 0
## A.status A.status -4.120076e-02 0
## A.start A.start -4.140127e-02 0
## A.incid A.incid -4.227791e-02 0
## A.determin A.determin -4.239248e-02 0
## A.posit A.posit -4.317584e-02 0
## A.progesteron A.progesteron -4.318635e-02 0
## A.aromatas A.aromatas -4.322202e-02 0
## A.local A.local -4.366301e-02 0
## A.main A.main -4.489330e-02 0
## A.present A.present -4.695298e-02 0
## A.aim A.aim -4.785872e-02 0
## A.possibl A.possibl -4.794828e-02 0
## A.perform A.perform -4.866672e-02 0
## A.larg A.larg -4.945536e-02 0
## A.shown A.shown -4.986612e-02 0
## A.hormon A.hormon -5.048861e-02 0
## A.period A.period -5.074228e-02 0
## A.reduct A.reduct -5.075505e-02 0
## T.clinic T.clinic -5.148394e-02 0
## A.second A.second -5.198940e-02 0
## A.breast A.breast -5.233490e-02 0
## A.estrogen A.estrogen -5.357349e-02 0
## A.cancer A.cancer -5.581714e-02 0
## A.tissu A.tissu -5.595253e-02 0
## A.adjuv A.adjuv -5.608234e-02 0
## A.case A.case -5.636252e-02 0
## A.find A.find -5.637269e-02 0
## A.obtain A.obtain -5.756790e-02 0
## A.suggest A.suggest -5.757780e-02 0
## A.within A.within -5.934658e-02 0
## A.remain A.remain -6.292028e-02 0
## A.design A.design -6.374264e-02 0
## A.X0001 A.X0001 -6.383307e-02 0
## A.potenti A.potenti -6.439343e-02 0
## A.import A.import -6.500431e-02 0
## A.detect A.detect -6.580512e-02 0
## A.women A.women -6.706568e-02 0
## A.recurr A.recurr -6.713675e-02 0
## A.investig A.investig -6.814008e-02 0
## A.chemotherapi A.chemotherapi -6.848646e-02 0
## A.histolog A.histolog -6.904065e-02 0
## A.lower A.lower -6.958213e-02 0
## T.earli T.earli -7.039691e-02 0
## A.select A.select -7.095647e-02 0
## A.first A.first -7.206416e-02 0
## A.signific A.signific -7.475443e-02 0
## A.low A.low -7.478252e-02 0
## A.mean A.mean -7.485297e-02 0
## T.women T.women -7.520353e-02 0
## A.radiotherapi A.radiotherapi -7.542001e-02 0
## A.clinic A.clinic -7.577756e-02 0
## A.also A.also -7.612550e-02 0
## A.relat A.relat -7.729160e-02 0
## A.life A.life -7.741262e-02 0
## A.size A.size -7.746176e-02 0
## A.effect A.effect -7.906464e-02 0
## A.outcom A.outcom -8.083204e-02 0
## A.wherea A.wherea -8.264291e-02 0
## A.may A.may -8.289575e-02 0
## A.indic A.indic -8.330481e-02 0
## A.system A.system -8.347057e-02 0
## A.test A.test -8.422639e-02 0
## A.metastas A.metastas -8.500380e-02 0
## A.type A.type -8.602559e-02 0
## A.function. A.function. -8.739253e-02 0
## A.analysi A.analysi -8.745562e-02 0
## A.avail A.avail -8.834244e-02 0
## A.negat A.negat -8.863047e-02 0
## A.analys A.analys -8.879582e-02 0
## A.X005 A.X005 -8.880757e-02 0
## A.report A.report -8.926696e-02 0
## A.show A.show -9.265780e-02 0
## A.provid A.provid -9.854909e-02 0
## A.tumor A.tumor -9.949416e-02 0
## A.qualiti A.qualiti -1.007081e-01 0
## A.can A.can -1.009582e-01 0
## A.doubleblind A.doubleblind -1.009869e-01 0
## A.tumour A.tumour -1.029714e-01 0
## A.serum A.serum -1.037758e-01 0
## A.factor A.factor -1.049804e-01 0
## A.analyz A.analyz -1.060879e-01 0
## A.proport A.proport -1.076947e-01 0
## A.valu A.valu -1.111873e-01 0
## A.sampl A.sampl -1.119119e-01 0
## A.inform A.inform -1.140931e-01 0
## A.assess A.assess -1.157562e-01 0
## A.need A.need -1.172494e-01 0
## A.independ A.independ -1.181013e-01 0
## A.correl A.correl -1.185756e-01 0
## A.increas A.increas -1.192067e-01 0
## A.level A.level -1.206844e-01 0
## A.placebo A.placebo -1.229875e-01 0
## A.multivari A.multivari -1.231479e-01 0
## A.model A.model -1.234019e-01 0
## A.trial A.trial -1.258125e-01 0
## A.high A.high -1.285034e-01 0
## A.score A.score -1.320092e-01 0
## A.express A.express -1.321424e-01 0
## A.found A.found -1.328394e-01 0
## A.particip A.particip -1.330350e-01 0
## A.control A.control -1.360290e-01 0
## A.examin A.examin -1.361521e-01 0
## A.prognost A.prognost -1.363777e-01 0
## A.marker A.marker -1.391338e-01 0
## A.reduc A.reduc -1.472907e-01 0
## A.cell A.cell -1.478798e-01 0
## A.measur A.measur -1.483961e-01 0
## A.chang A.chang -1.494941e-01 0
## T.respons T.respons -1.496979e-01 0
## A.baselin A.baselin -1.517685e-01 0
## A.identifi A.identifi -1.524843e-01 0
## A.decreas A.decreas -1.536759e-01 0
## A.bone A.bone -1.552270e-01 0
## A.prevent A.prevent -1.556778e-01 0
## A.data A.data -1.605586e-01 0
## A.associ A.associ -1.738833e-01 0
## A.predict A.predict -1.756009e-01 0
## T.effect T.effect -1.798643e-01 0
## A.risk A.risk -1.845209e-01 0
## A.use A.use -2.345315e-01 0
## T.has.http T.has.http NA 0
## cor.y.abs cor.high.X is.ConditionalX.y
## trial 1.000000e+00 <NA> NA
## T.phase 4.407918e-01 <NA> TRUE
## A.toxic 3.546830e-01 <NA> TRUE
## A.mgm2 3.476197e-01 <NA> TRUE
## T.metastat 3.049464e-01 <NA> TRUE
## A.everi 2.932475e-01 <NA> TRUE
## A.median 2.890542e-01 <NA> TRUE
## A.rate 2.658285e-01 <NA> TRUE
## T.studi 2.657631e-01 <NA> TRUE
## T.versus 2.497154e-01 <NA> TRUE
## T.advanc 2.469661e-01 <NA> TRUE
## A.respons 2.450234e-01 <NA> TRUE
## A.metastat 2.417899e-01 <NA> TRUE
## A.progress 2.284989e-01 <NA> TRUE
## A.advanc 2.281995e-01 <NA> TRUE
## A.neutropenia 2.262877e-01 <NA> TRUE
## A.partial 2.185387e-01 <NA> TRUE
## A.toler 2.093203e-01 <NA> TRUE
## A.combin 2.088035e-01 <NA> TRUE
## A.firstlin 2.039271e-01 <NA> TRUE
## A.regimen 2.028112e-01 <NA> TRUE
## T.breast 1.989942e-01 T.cancer TRUE
## A.cyclophosphamid 1.958128e-01 <NA> TRUE
## T.docetaxel 1.947844e-01 <NA> TRUE
## T.cyclophosphamid 1.932487e-01 <NA> TRUE
## A.overal 1.927655e-01 <NA> TRUE
## A.week 1.918602e-01 <NA> TRUE
## A.month 1.887787e-01 <NA> TRUE
## A.surviv 1.844165e-01 <NA> TRUE
## T.combin 1.835246e-01 <NA> TRUE
## A.day 1.770836e-01 <NA> TRUE
## A.arm 1.729602e-01 <NA> TRUE
## A.docetaxel 1.721916e-01 <NA> TRUE
## T.cancer 1.680898e-01 <NA> TRUE
## T.iii 1.663443e-01 <NA> TRUE
## T.compar 1.658208e-01 <NA> TRUE
## A.durat 1.644884e-01 <NA> TRUE
## T.plus 1.614591e-01 <NA> TRUE
## A.given 1.577815e-01 <NA> TRUE
## A.epirubicin 1.576660e-01 <NA> TRUE
## A.mbc 1.540985e-01 <NA> TRUE
## A.previous 1.535045e-01 <NA> TRUE
## A.X100 1.513446e-01 <NA> TRUE
## A.four 1.512749e-01 <NA> TRUE
## A.cycl 1.511902e-01 <NA> TRUE
## A.methotrex 1.508929e-01 <NA> TRUE
## T.group 1.488481e-01 <NA> TRUE
## A.phase 1.481848e-01 <NA> TRUE
## A.diseas 1.454492e-01 <NA> TRUE
## A.efficaci 1.446412e-01 <NA> TRUE
## A.X500 1.433281e-01 <NA> TRUE
## A.grade 1.432705e-01 <NA> TRUE
## A.patient 1.420881e-01 <NA> TRUE
## A.occur 1.363943e-01 <NA> TRUE
## A.hundr 1.362355e-01 <NA> TRUE
## A.schedul 1.358289e-01 <NA> TRUE
## T.doxorubicin 1.324163e-01 <NA> TRUE
## A.either 1.309481e-01 <NA> TRUE
## A.dose 1.284812e-01 <NA> TRUE
## A.activ 1.271691e-01 <NA> TRUE
## A.progressionfre 1.270990e-01 <NA> TRUE
## A.safeti 1.217685e-01 <NA> TRUE
## A.doxorubicin 1.217472e-01 <NA> TRUE
## A.seen 1.203577e-01 <NA> TRUE
## A.elig 1.201664e-01 <NA> TRUE
## T.trial 1.195178e-01 <NA> TRUE
## A.hematolog 1.187144e-01 <NA> TRUE
## A.plus 1.174459e-01 <NA> TRUE
## A.vomit 1.161292e-01 A.nausea TRUE
## A.respect 1.158080e-01 <NA> TRUE
## A.fluorouracil 1.156599e-01 <NA> TRUE
## A.X5fluorouracil 1.153122e-01 <NA> TRUE
## A.superior 1.144624e-01 <NA> TRUE
## A.enter 1.130515e-01 <NA> TRUE
## T.num.chars 1.122127e-01 T.num.words TRUE
## A.two 1.118305e-01 <NA> TRUE
## A.object 1.065300e-01 <NA> TRUE
## A.rang 1.050787e-01 <NA> TRUE
## A.paclitaxel 1.044681e-01 <NA> TRUE
## T.num.words 1.040679e-01 T.num.words.unq TRUE
## A.three 1.031348e-01 <NA> TRUE
## T.random 1.028374e-01 <NA> TRUE
## A.prior 1.026878e-01 <NA> TRUE
## A.cmf 1.023838e-01 <NA> TRUE
## A.sequenti 9.916268e-02 <NA> TRUE
## A.oral 9.728974e-02 <NA> TRUE
## A.complet 9.717265e-02 <NA> TRUE
## A.diseasefre 9.577232e-02 <NA> TRUE
## A.achiev 9.514020e-02 <NA> TRUE
## A.similar 9.371701e-02 <NA> TRUE
## T.randomis 9.370110e-02 <NA> TRUE
## A.six 9.362652e-02 <NA> TRUE
## A.secondari 8.944036e-02 <NA> TRUE
## A.well 8.870282e-02 <NA> TRUE
## T.num.words.unq 8.542797e-02 <NA> TRUE
## A.event 8.489612e-02 <NA> TRUE
## A.anthracyclin 8.474734e-02 <NA> TRUE
## A.death 8.448898e-02 <NA> TRUE
## A.longer 8.421948e-02 <NA> TRUE
## A.eight 8.362672e-02 <NA> TRUE
## A.one 8.221967e-02 <NA> TRUE
## A.enrol 7.928104e-02 <NA> TRUE
## A.term 7.818056e-02 <NA> TRUE
## A.administ 7.791871e-02 <NA> TRUE
## A.failur 7.764819e-02 <NA> TRUE
## A.common 7.737623e-02 <NA> TRUE
## A.frequent 7.708184e-02 <NA> TRUE
## A.assign 7.540280e-02 <NA> TRUE
## A.nodeposit 7.498902e-02 <NA> TRUE
## A.respond 7.372052e-02 <NA> TRUE
## A.endpoint 7.093389e-02 <NA> TRUE
## A.stabl 7.053656e-02 <NA> TRUE
## A.confid 7.049294e-02 <NA> TRUE
## A.prolong 6.968693e-02 <NA> TRUE
## A.continu 6.889854e-02 <NA> TRUE
## A.daili 6.879414e-02 <NA> TRUE
## A.experienc 6.719543e-02 <NA> TRUE
## A.follow 6.707377e-02 <NA> TRUE
## A.treatment 6.578098e-02 <NA> TRUE
## A.end 6.542423e-02 A.point TRUE
## A.time 6.178757e-02 <NA> TRUE
## A.receiv 6.090943e-02 <NA> TRUE
## A.dfs 6.076035e-02 <NA> TRUE
## A.primari 6.020830e-02 <NA> TRUE
## A.interv 5.928858e-02 <NA> TRUE
## A.compar 5.923358e-02 <NA> TRUE
## A.total 5.916846e-02 <NA> TRUE
## T.treatment 5.769991e-02 <NA> TRUE
## A.intraven 5.564821e-02 <NA> TRUE
## T.tamoxifen 5.564564e-02 <NA> TRUE
## T.result 5.427221e-02 <NA> TRUE
## A.cours 5.392045e-02 <NA> TRUE
## A.observ 5.273205e-02 <NA> TRUE
## A.advers 5.167782e-02 <NA> TRUE
## A.evalu 5.098452e-02 <NA> TRUE
## A.versus 5.026899e-02 <NA> TRUE
## A.pretreat 4.965851e-02 <NA> TRUE
## A.point 4.946137e-02 <NA> TRUE
## T.postmenopaus 4.801689e-02 <NA> TRUE
## A.includ 4.584644e-02 <NA> TRUE
## A.seven 4.541994e-02 <NA> TRUE
## A.nausea 4.537254e-02 <NA> TRUE
## .rnorm 4.466651e-02 <NA> TRUE
## A.five 4.460546e-02 <NA> TRUE
## A.better 4.133124e-02 <NA> TRUE
## A.multicent 4.053909e-02 <NA> TRUE
## A.oper 3.993046e-02 <NA> TRUE
## A.stage 3.975581e-02 <NA> TRUE
## A.alon 3.907515e-02 <NA> TRUE
## A.less 3.732604e-02 <NA> TRUE
## A.random 3.663241e-02 <NA> TRUE
## A.although 3.604544e-02 <NA> TRUE
## A.addit 3.510086e-02 <NA> TRUE
## A.agent 3.257901e-02 <NA> TRUE
## A.due 3.164831e-02 <NA> TRUE
## A.andor 3.086932e-02 <NA> TRUE
## A.lymph 3.005185e-02 <NA> TRUE
## A.followup 2.973279e-02 <NA> TRUE
## A.postmenopaus 2.765137e-02 <NA> TRUE
## A.involv 2.700336e-02 <NA> TRUE
## A.side 2.689790e-02 <NA> TRUE
## A.standard 2.641969e-02 <NA> TRUE
## A.receptorposit 2.492618e-02 <NA> TRUE
## A.conduct 2.467401e-02 <NA> TRUE
## A.neoadjuv 2.210203e-02 <NA> TRUE
## A.major 2.209530e-02 <NA> TRUE
## A.administr 2.040770e-02 <NA> TRUE
## A.limit 1.884501e-02 <NA> TRUE
## A.consist 1.855504e-02 <NA> TRUE
## A.differ 1.797677e-02 <NA> TRUE
## A.num.words.unq 1.784364e-02 <NA> TRUE
## A.singl 1.695022e-02 <NA> TRUE
## A.hazard 1.634076e-02 <NA> TRUE
## A.sever 1.516401e-02 <NA> TRUE
## A.result 1.442596e-02 <NA> TRUE
## A.randomis 1.425652e-02 <NA> TRUE
## A.appear 1.314392e-02 <NA> TRUE
## A.treat 1.103432e-02 <NA> TRUE
## A.profil 1.092909e-02 <NA> TRUE
## A.year 9.566746e-03 <NA> TRUE
## A.accord 9.173456e-03 <NA> TRUE
## T.patient 7.667604e-03 <NA> TRUE
## A.site 7.553580e-03 <NA> TRUE
## A.improv 7.017871e-03 <NA> TRUE
## A.tamoxifen 7.004532e-03 <NA> TRUE
## A.ratio 6.872134e-03 <NA> TRUE
## A.evid 5.914321e-03 <NA> TRUE
## A.iii 3.062659e-03 <NA> TRUE
## A.conclus 2.541412e-03 <NA> TRUE
## A.demonstr 1.364004e-03 <NA> TRUE
## A.statist 1.141908e-03 <NA> TRUE
## A.among 3.632848e-04 <NA> TRUE
## A.relaps 7.929870e-06 <NA> TRUE
## A.human 9.121557e-04 <NA> TRUE
## A.without 1.584596e-03 <NA> TRUE
## A.surgeri 2.256826e-03 <NA> TRUE
## A.drug 2.504858e-03 <NA> TRUE
## A.infus 2.691231e-03 <NA> TRUE
## T.therapi 3.064253e-03 <NA> TRUE
## A.higher 3.196262e-03 <NA> TRUE
## A.purpos 6.501960e-03 <NA> TRUE
## A.premenopaus 6.600118e-03 <NA> TRUE
## A.num.chars 7.349760e-03 <NA> TRUE
## A.background 9.175455e-03 <NA> TRUE
## A.benefit 9.524764e-03 <NA> TRUE
## A.growth 9.887605e-03 <NA> TRUE
## T.chemotherapi 9.909758e-03 <NA> TRUE
## A.confirm 1.221603e-02 <NA> TRUE
## A.inhibitor 1.239898e-02 <NA> TRUE
## A.characterist 1.258143e-02 <NA> TRUE
## A.whether 1.291745e-02 <NA> TRUE
## A.popul 1.296710e-02 <NA> TRUE
## A.initi 1.332423e-02 <NA> TRUE
## A.group 1.337887e-02 <NA> TRUE
## A.mastectomi 1.359074e-02 <NA> TRUE
## A.least 1.387475e-02 <NA> TRUE
## A.consid 1.417986e-02 <NA> TRUE
## A.node 1.504773e-02 <NA> TRUE
## A.axillari 1.556611e-02 <NA> TRUE
## A.num.words 1.799787e-02 <NA> TRUE
## A.caus 1.915695e-02 <NA> TRUE
## A.endocrin 1.936997e-02 <NA> TRUE
## A.estim 1.960142e-02 <NA> TRUE
## A.new 2.033234e-02 <NA> TRUE
## A.postop 2.083001e-02 <NA> TRUE
## A.distant 2.186214e-02 <NA> TRUE
## A.regard 2.212204e-02 <NA> TRUE
## A.requir 2.281593e-02 <NA> TRUE
## A.defin 2.332432e-02 <NA> TRUE
## A.trend 2.337564e-02 <NA> TRUE
## A.subgroup 2.338162e-02 <NA> TRUE
## A.receptor 2.389969e-02 <NA> TRUE
## A.greater 2.407688e-02 <NA> TRUE
## A.therapi 2.549221e-02 <NA> TRUE
## A.studi 2.594717e-02 <NA> TRUE
## A.howev 2.671381e-02 <NA> TRUE
## A.support 2.793158e-02 <NA> TRUE
## T.adjuv 2.813822e-02 <NA> TRUE
## A.number 2.963870e-02 <NA> TRUE
## A.develop 3.000773e-02 <NA> TRUE
## A.comparison 3.040785e-02 <NA> TRUE
## A.prospect 3.102924e-02 <NA> TRUE
## A.regress 3.139311e-02 <NA> TRUE
## A.per 3.173574e-02 <NA> TRUE
## A.age 3.188645e-02 <NA> TRUE
## A.general 3.234203e-02 <NA> TRUE
## A.earli 3.276589e-02 <NA> TRUE
## A.X001 3.311267e-02 <NA> TRUE
## A.base 3.357553e-02 <NA> TRUE
## A.has.http 3.472002e-02 <NA> FALSE
## A.carcinoma 3.493790e-02 <NA> TRUE
## A.method 3.582004e-02 <NA> TRUE
## A.patholog 3.751957e-02 <NA> TRUE
## A.set 3.997140e-02 <NA> TRUE
## A.her2 4.085805e-02 <NA> TRUE
## A.status 4.120076e-02 <NA> TRUE
## A.start 4.140127e-02 <NA> TRUE
## A.incid 4.227791e-02 <NA> TRUE
## A.determin 4.239248e-02 <NA> TRUE
## A.posit 4.317584e-02 <NA> TRUE
## A.progesteron 4.318635e-02 <NA> TRUE
## A.aromatas 4.322202e-02 <NA> TRUE
## A.local 4.366301e-02 <NA> TRUE
## A.main 4.489330e-02 <NA> TRUE
## A.present 4.695298e-02 <NA> TRUE
## A.aim 4.785872e-02 <NA> TRUE
## A.possibl 4.794828e-02 <NA> TRUE
## A.perform 4.866672e-02 <NA> TRUE
## A.larg 4.945536e-02 <NA> TRUE
## A.shown 4.986612e-02 <NA> TRUE
## A.hormon 5.048861e-02 <NA> TRUE
## A.period 5.074228e-02 <NA> TRUE
## A.reduct 5.075505e-02 <NA> TRUE
## T.clinic 5.148394e-02 <NA> TRUE
## A.second 5.198940e-02 <NA> TRUE
## A.breast 5.233490e-02 <NA> TRUE
## A.estrogen 5.357349e-02 <NA> TRUE
## A.cancer 5.581714e-02 A.breast TRUE
## A.tissu 5.595253e-02 <NA> TRUE
## A.adjuv 5.608234e-02 <NA> TRUE
## A.case 5.636252e-02 <NA> TRUE
## A.find 5.637269e-02 <NA> TRUE
## A.obtain 5.756790e-02 <NA> TRUE
## A.suggest 5.757780e-02 <NA> TRUE
## A.within 5.934658e-02 <NA> TRUE
## A.remain 6.292028e-02 <NA> TRUE
## A.design 6.374264e-02 <NA> TRUE
## A.X0001 6.383307e-02 <NA> TRUE
## A.potenti 6.439343e-02 <NA> TRUE
## A.import 6.500431e-02 <NA> TRUE
## A.detect 6.580512e-02 <NA> TRUE
## A.women 6.706568e-02 <NA> TRUE
## A.recurr 6.713675e-02 <NA> TRUE
## A.investig 6.814008e-02 <NA> TRUE
## A.chemotherapi 6.848646e-02 <NA> TRUE
## A.histolog 6.904065e-02 <NA> TRUE
## A.lower 6.958213e-02 <NA> TRUE
## T.earli 7.039691e-02 <NA> TRUE
## A.select 7.095647e-02 <NA> TRUE
## A.first 7.206416e-02 <NA> TRUE
## A.signific 7.475443e-02 <NA> TRUE
## A.low 7.478252e-02 <NA> TRUE
## A.mean 7.485297e-02 <NA> TRUE
## T.women 7.520353e-02 <NA> TRUE
## A.radiotherapi 7.542001e-02 <NA> TRUE
## A.clinic 7.577756e-02 <NA> TRUE
## A.also 7.612550e-02 <NA> TRUE
## A.relat 7.729160e-02 <NA> TRUE
## A.life 7.741262e-02 <NA> TRUE
## A.size 7.746176e-02 <NA> TRUE
## A.effect 7.906464e-02 <NA> TRUE
## A.outcom 8.083204e-02 <NA> TRUE
## A.wherea 8.264291e-02 <NA> TRUE
## A.may 8.289575e-02 <NA> TRUE
## A.indic 8.330481e-02 <NA> TRUE
## A.system 8.347057e-02 <NA> TRUE
## A.test 8.422639e-02 <NA> TRUE
## A.metastas 8.500380e-02 <NA> TRUE
## A.type 8.602559e-02 <NA> TRUE
## A.function. 8.739253e-02 <NA> TRUE
## A.analysi 8.745562e-02 <NA> TRUE
## A.avail 8.834244e-02 <NA> TRUE
## A.negat 8.863047e-02 <NA> TRUE
## A.analys 8.879582e-02 <NA> TRUE
## A.X005 8.880757e-02 <NA> TRUE
## A.report 8.926696e-02 <NA> TRUE
## A.show 9.265780e-02 <NA> TRUE
## A.provid 9.854909e-02 <NA> TRUE
## A.tumor 9.949416e-02 <NA> TRUE
## A.qualiti 1.007081e-01 A.life TRUE
## A.can 1.009582e-01 <NA> TRUE
## A.doubleblind 1.009869e-01 <NA> TRUE
## A.tumour 1.029714e-01 <NA> TRUE
## A.serum 1.037758e-01 <NA> TRUE
## A.factor 1.049804e-01 <NA> TRUE
## A.analyz 1.060879e-01 <NA> TRUE
## A.proport 1.076947e-01 <NA> TRUE
## A.valu 1.111873e-01 <NA> TRUE
## A.sampl 1.119119e-01 <NA> TRUE
## A.inform 1.140931e-01 <NA> TRUE
## A.assess 1.157562e-01 <NA> TRUE
## A.need 1.172494e-01 <NA> TRUE
## A.independ 1.181013e-01 <NA> TRUE
## A.correl 1.185756e-01 <NA> TRUE
## A.increas 1.192067e-01 <NA> TRUE
## A.level 1.206844e-01 <NA> TRUE
## A.placebo 1.229875e-01 <NA> TRUE
## A.multivari 1.231479e-01 <NA> TRUE
## A.model 1.234019e-01 <NA> TRUE
## A.trial 1.258125e-01 <NA> TRUE
## A.high 1.285034e-01 <NA> TRUE
## A.score 1.320092e-01 <NA> TRUE
## A.express 1.321424e-01 <NA> TRUE
## A.found 1.328394e-01 <NA> TRUE
## A.particip 1.330350e-01 <NA> TRUE
## A.control 1.360290e-01 <NA> TRUE
## A.examin 1.361521e-01 <NA> TRUE
## A.prognost 1.363777e-01 <NA> TRUE
## A.marker 1.391338e-01 <NA> TRUE
## A.reduc 1.472907e-01 <NA> TRUE
## A.cell 1.478798e-01 <NA> TRUE
## A.measur 1.483961e-01 <NA> TRUE
## A.chang 1.494941e-01 <NA> TRUE
## T.respons 1.496979e-01 <NA> TRUE
## A.baselin 1.517685e-01 <NA> TRUE
## A.identifi 1.524843e-01 <NA> TRUE
## A.decreas 1.536759e-01 <NA> TRUE
## A.bone 1.552270e-01 <NA> TRUE
## A.prevent 1.556778e-01 <NA> TRUE
## A.data 1.605586e-01 <NA> TRUE
## A.associ 1.738833e-01 <NA> TRUE
## A.predict 1.756009e-01 <NA> TRUE
## T.effect 1.798643e-01 <NA> TRUE
## A.risk 1.845209e-01 <NA> TRUE
## A.use 2.345315e-01 <NA> TRUE
## T.has.http NA <NA> FALSE
## is.cor.y.abs.low
## trial FALSE
## T.phase FALSE
## A.toxic FALSE
## A.mgm2 FALSE
## T.metastat FALSE
## A.everi FALSE
## A.median FALSE
## A.rate FALSE
## T.studi FALSE
## T.versus FALSE
## T.advanc FALSE
## A.respons FALSE
## A.metastat FALSE
## A.progress FALSE
## A.advanc FALSE
## A.neutropenia FALSE
## A.partial FALSE
## A.toler FALSE
## A.combin FALSE
## A.firstlin FALSE
## A.regimen FALSE
## T.breast FALSE
## A.cyclophosphamid FALSE
## T.docetaxel FALSE
## T.cyclophosphamid FALSE
## A.overal FALSE
## A.week FALSE
## A.month FALSE
## A.surviv FALSE
## T.combin FALSE
## A.day FALSE
## A.arm FALSE
## A.docetaxel FALSE
## T.cancer FALSE
## T.iii FALSE
## T.compar FALSE
## A.durat FALSE
## T.plus FALSE
## A.given FALSE
## A.epirubicin FALSE
## A.mbc FALSE
## A.previous FALSE
## A.X100 FALSE
## A.four FALSE
## A.cycl FALSE
## A.methotrex FALSE
## T.group FALSE
## A.phase FALSE
## A.diseas FALSE
## A.efficaci FALSE
## A.X500 FALSE
## A.grade FALSE
## A.patient FALSE
## A.occur FALSE
## A.hundr FALSE
## A.schedul FALSE
## T.doxorubicin FALSE
## A.either FALSE
## A.dose FALSE
## A.activ FALSE
## A.progressionfre FALSE
## A.safeti FALSE
## A.doxorubicin FALSE
## A.seen FALSE
## A.elig FALSE
## T.trial FALSE
## A.hematolog FALSE
## A.plus FALSE
## A.vomit FALSE
## A.respect FALSE
## A.fluorouracil FALSE
## A.X5fluorouracil FALSE
## A.superior FALSE
## A.enter FALSE
## T.num.chars FALSE
## A.two FALSE
## A.object FALSE
## A.rang FALSE
## A.paclitaxel FALSE
## T.num.words FALSE
## A.three FALSE
## T.random FALSE
## A.prior FALSE
## A.cmf FALSE
## A.sequenti FALSE
## A.oral FALSE
## A.complet FALSE
## A.diseasefre FALSE
## A.achiev FALSE
## A.similar FALSE
## T.randomis FALSE
## A.six FALSE
## A.secondari FALSE
## A.well FALSE
## T.num.words.unq FALSE
## A.event FALSE
## A.anthracyclin FALSE
## A.death FALSE
## A.longer FALSE
## A.eight FALSE
## A.one FALSE
## A.enrol FALSE
## A.term FALSE
## A.administ FALSE
## A.failur FALSE
## A.common FALSE
## A.frequent FALSE
## A.assign FALSE
## A.nodeposit FALSE
## A.respond FALSE
## A.endpoint FALSE
## A.stabl FALSE
## A.confid FALSE
## A.prolong FALSE
## A.continu FALSE
## A.daili FALSE
## A.experienc FALSE
## A.follow FALSE
## A.treatment FALSE
## A.end FALSE
## A.time FALSE
## A.receiv FALSE
## A.dfs FALSE
## A.primari FALSE
## A.interv FALSE
## A.compar FALSE
## A.total FALSE
## T.treatment FALSE
## A.intraven FALSE
## T.tamoxifen FALSE
## T.result FALSE
## A.cours FALSE
## A.observ FALSE
## A.advers FALSE
## A.evalu FALSE
## A.versus FALSE
## A.pretreat FALSE
## A.point FALSE
## T.postmenopaus FALSE
## A.includ FALSE
## A.seven FALSE
## A.nausea FALSE
## .rnorm FALSE
## A.five TRUE
## A.better TRUE
## A.multicent TRUE
## A.oper TRUE
## A.stage TRUE
## A.alon TRUE
## A.less TRUE
## A.random TRUE
## A.although TRUE
## A.addit TRUE
## A.agent TRUE
## A.due TRUE
## A.andor TRUE
## A.lymph TRUE
## A.followup TRUE
## A.postmenopaus TRUE
## A.involv TRUE
## A.side TRUE
## A.standard TRUE
## A.receptorposit TRUE
## A.conduct TRUE
## A.neoadjuv TRUE
## A.major TRUE
## A.administr TRUE
## A.limit TRUE
## A.consist TRUE
## A.differ TRUE
## A.num.words.unq TRUE
## A.singl TRUE
## A.hazard TRUE
## A.sever TRUE
## A.result TRUE
## A.randomis TRUE
## A.appear TRUE
## A.treat TRUE
## A.profil TRUE
## A.year TRUE
## A.accord TRUE
## T.patient TRUE
## A.site TRUE
## A.improv TRUE
## A.tamoxifen TRUE
## A.ratio TRUE
## A.evid TRUE
## A.iii TRUE
## A.conclus TRUE
## A.demonstr TRUE
## A.statist TRUE
## A.among TRUE
## A.relaps TRUE
## A.human TRUE
## A.without TRUE
## A.surgeri TRUE
## A.drug TRUE
## A.infus TRUE
## T.therapi TRUE
## A.higher TRUE
## A.purpos TRUE
## A.premenopaus TRUE
## A.num.chars TRUE
## A.background TRUE
## A.benefit TRUE
## A.growth TRUE
## T.chemotherapi TRUE
## A.confirm TRUE
## A.inhibitor TRUE
## A.characterist TRUE
## A.whether TRUE
## A.popul TRUE
## A.initi TRUE
## A.group TRUE
## A.mastectomi TRUE
## A.least TRUE
## A.consid TRUE
## A.node TRUE
## A.axillari TRUE
## A.num.words TRUE
## A.caus TRUE
## A.endocrin TRUE
## A.estim TRUE
## A.new TRUE
## A.postop TRUE
## A.distant TRUE
## A.regard TRUE
## A.requir TRUE
## A.defin TRUE
## A.trend TRUE
## A.subgroup TRUE
## A.receptor TRUE
## A.greater TRUE
## A.therapi TRUE
## A.studi TRUE
## A.howev TRUE
## A.support TRUE
## T.adjuv TRUE
## A.number TRUE
## A.develop TRUE
## A.comparison TRUE
## A.prospect TRUE
## A.regress TRUE
## A.per TRUE
## A.age TRUE
## A.general TRUE
## A.earli TRUE
## A.X001 TRUE
## A.base TRUE
## A.has.http TRUE
## A.carcinoma TRUE
## A.method TRUE
## A.patholog TRUE
## A.set TRUE
## A.her2 TRUE
## A.status TRUE
## A.start TRUE
## A.incid TRUE
## A.determin TRUE
## A.posit TRUE
## A.progesteron TRUE
## A.aromatas TRUE
## A.local TRUE
## A.main FALSE
## A.present FALSE
## A.aim FALSE
## A.possibl FALSE
## A.perform FALSE
## A.larg FALSE
## A.shown FALSE
## A.hormon FALSE
## A.period FALSE
## A.reduct FALSE
## T.clinic FALSE
## A.second FALSE
## A.breast FALSE
## A.estrogen FALSE
## A.cancer FALSE
## A.tissu FALSE
## A.adjuv FALSE
## A.case FALSE
## A.find FALSE
## A.obtain FALSE
## A.suggest FALSE
## A.within FALSE
## A.remain FALSE
## A.design FALSE
## A.X0001 FALSE
## A.potenti FALSE
## A.import FALSE
## A.detect FALSE
## A.women FALSE
## A.recurr FALSE
## A.investig FALSE
## A.chemotherapi FALSE
## A.histolog FALSE
## A.lower FALSE
## T.earli FALSE
## A.select FALSE
## A.first FALSE
## A.signific FALSE
## A.low FALSE
## A.mean FALSE
## T.women FALSE
## A.radiotherapi FALSE
## A.clinic FALSE
## A.also FALSE
## A.relat FALSE
## A.life FALSE
## A.size FALSE
## A.effect FALSE
## A.outcom FALSE
## A.wherea FALSE
## A.may FALSE
## A.indic FALSE
## A.system FALSE
## A.test FALSE
## A.metastas FALSE
## A.type FALSE
## A.function. FALSE
## A.analysi FALSE
## A.avail FALSE
## A.negat FALSE
## A.analys FALSE
## A.X005 FALSE
## A.report FALSE
## A.show FALSE
## A.provid FALSE
## A.tumor FALSE
## A.qualiti FALSE
## A.can FALSE
## A.doubleblind FALSE
## A.tumour FALSE
## A.serum FALSE
## A.factor FALSE
## A.analyz FALSE
## A.proport FALSE
## A.valu FALSE
## A.sampl FALSE
## A.inform FALSE
## A.assess FALSE
## A.need FALSE
## A.independ FALSE
## A.correl FALSE
## A.increas FALSE
## A.level FALSE
## A.placebo FALSE
## A.multivari FALSE
## A.model FALSE
## A.trial FALSE
## A.high FALSE
## A.score FALSE
## A.express FALSE
## A.found FALSE
## A.particip FALSE
## A.control FALSE
## A.examin FALSE
## A.prognost FALSE
## A.marker FALSE
## A.reduc FALSE
## A.cell FALSE
## A.measur FALSE
## A.chang FALSE
## T.respons FALSE
## A.baselin FALSE
## A.identifi FALSE
## A.decreas FALSE
## A.bone FALSE
## A.prevent FALSE
## A.data FALSE
## A.associ FALSE
## A.predict FALSE
## T.effect FALSE
## A.risk FALSE
## A.use FALSE
## T.has.http NAglb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.models",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor
## elapsed7 remove_correlated_features 4 1
## elapsed8 fit.models 5 0
## elapsed
## elapsed7 27.537
## elapsed8 40.403Step 5: fit models
glb_models_lst <- list(); glb_models_df <- data.frame()
if (glb_is_classification && glb_is_binomial &&
(length(unique(glb_trnent_df[, glb_rsp_var])) < 2))
stop("glb_trnent_df$", glb_rsp_var, ": contains less than 2 unique values: ",
paste0(unique(glb_trnent_df[, glb_rsp_var]), collapse=", "))
max_cor_y_x_var <- orderBy(~ -cor.y.abs,
subset(glb_feats_df, (exclude.as.feat == 0) & !is.cor.y.abs.low))[1, "id"]
if (!is.null(glb_Baseline_mdl_var)) {
if ((max_cor_y_x_var != glb_Baseline_mdl_var) &
(glb_feats_df[max_cor_y_x_var, "cor.y.abs"] >
glb_feats_df[glb_Baseline_mdl_var, "cor.y.abs"]))
stop(max_cor_y_x_var, " has a lower correlation with ", glb_rsp_var,
" than the Baseline var: ", glb_Baseline_mdl_var)
}
glb_model_type <- ifelse(glb_is_regression, "regression", "classification")
# Baseline
if (!is.null(glb_Baseline_mdl_var))
ret_lst <- myfit_mdl_fn(model_id="Baseline", model_method="mybaseln_classfr",
indep_vars_vctr=glb_Baseline_mdl_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)
# Most Frequent Outcome "MFO" model: mean(y) for regression
# Not using caret's nullModel since model stats not avl
# Cannot use rpart for multinomial classification since it predicts non-MFO
ret_lst <- myfit_mdl(model_id="MFO",
model_method=ifelse(glb_is_regression, "lm", "myMFO_classfr"),
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)## [1] "fitting model: MFO.myMFO_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## [1] "in MFO.Classifier$fit"
## [1] "unique.vals:"
## [1] N Y
## Levels: N Y
## [1] "unique.prob:"
## y
## N Y
## 0.5606759 0.4393241
## [1] "MFO.val:"
## [1] "N"
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 -none- numeric
## MFO.val 1 -none- character
## x.names 1 -none- character
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character## Loading required package: ROCR
## Loading required package: gplots
##
## Attaching package: 'gplots'
##
## The following object is masked from 'package:stats':
##
## lowess## [1] "in MFO.Classifier$predict"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.5606759 0.4393241
## 2 0.5606759 0.4393241
## 3 0.5606759 0.4393241
## 4 0.5606759 0.4393241
## 5 0.5606759 0.4393241
## 6 0.5606759 0.4393241
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.MFO.myMFO_classfr.N
## 1 N 730
## 2 Y 572
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.MFO.myMFO_classfr.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.MFO.myMFO_classfr.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 730 0
## Y 572 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.606759e-01 0.000000e+00 5.332221e-01 5.878552e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 5.115865e-01 5.609658e-126
## [1] "in MFO.Classifier$predict"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.5606759 0.4393241
## 2 0.5606759 0.4393241
## 3 0.5606759 0.4393241
## 4 0.5606759 0.4393241
## 5 0.5606759 0.4393241
## 6 0.5606759 0.4393241
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.MFO.myMFO_classfr.N
## 1 N 313
## 2 Y 245
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.MFO.myMFO_classfr.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.MFO.myMFO_classfr.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 313 0
## Y 245 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.609319e-01 0.000000e+00 5.186280e-01 6.025874e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 5.176975e-01 8.703586e-55
## model_id model_method feats max.nTuningRuns
## 1 MFO.myMFO_classfr myMFO_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.38 0.002 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.5606759
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.5332221 0.5878552 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.5609319
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.518628 0.6025874 0if (glb_is_classification)
# "random" model - only for classification;
# none needed for regression since it is same as MFO
ret_lst <- myfit_mdl(model_id="Random", model_method="myrandom_classfr",
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)## [1] "fitting model: Random.myrandom_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 table numeric
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] "in Random.Classifier$prob"
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 415 314
## Y 315 258
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 415
## 2 Y 314
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 315
## 2 258
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.6104589
## 3 0.2 0.6104589
## 4 0.3 0.6104589
## 5 0.4 0.6104589
## 6 0.5 0.4506550
## 7 0.6 0.0000000
## 8 0.7 0.0000000
## 9 0.8 0.0000000
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.Y
## 1 N 730
## 2 Y 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 730
## 2 572
## Prediction
## Reference N Y
## N 0 730
## Y 0 572
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.393241e-01 0.000000e+00 4.121448e-01 4.667779e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.436635e-160
## [1] "in Random.Classifier$prob"
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 177 140
## Y 136 105
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 177
## 2 Y 140
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 136
## 2 105
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6102117
## 3 0.2 0.6102117
## 4 0.3 0.6102117
## 5 0.4 0.6102117
## 6 0.5 0.4320988
## 7 0.6 0.0000000
## 8 0.7 0.0000000
## 9 0.8 0.0000000
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.Y
## 1 N 313
## 2 Y 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Random.myrandom_classfr.Y
## 1 313
## 2 245
## Prediction
## Reference N Y
## N 0 313
## Y 0 245
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.390681e-01 0.000000e+00 3.974126e-01 4.813720e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.320342e-69
## model_id model_method feats max.nTuningRuns
## 1 Random.myrandom_classfr myrandom_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.243 0.001 0.5097711
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.4 0.6104589 0.4393241
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.4121448 0.4667779 0 0.4970333
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.4 0.6102117 0.4390681
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.3974126 0.481372 0# Any models that have tuning parameters has "better" results with cross-validation
# (except rf) & "different" results for different outcome metrics
# Max.cor.Y
# Check impact of cv
# rpart is not a good candidate since caret does not optimize cp (only tuning parameter of rpart) well
ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)## [1] "fitting model: Max.cor.Y.cv.0.rpart"
## [1] " indep_vars: T.phase"## Loading required package: rpart## Fitting cp = 0.355 on full training set## Loading required package: rpart.plot
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1302
##
## CP nsplit rel error
## 1 0.3548951 0 1
##
## Node number 1: 1302 observations
## predicted class=N expected loss=0.4393241 P(node) =1
## class counts: 730 572
## probabilities: 0.561 0.439
##
## n= 1302
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1302 572 N (0.5606759 0.4393241) *
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 730
## 2 Y 572
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Max.cor.Y.cv.0.rpart.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 730 0
## Y 572 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.606759e-01 0.000000e+00 5.332221e-01 5.878552e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 5.115865e-01 5.609658e-126
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 313
## 2 Y 245
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Max.cor.Y.cv.0.rpart.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 313 0
## Y 245 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.609319e-01 0.000000e+00 5.186280e-01 6.025874e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 5.176975e-01 8.703586e-55
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.cv.0.rpart rpart T.phase 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.601 0.02 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.5606759
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.5332221 0.5878552 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.5609319
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.518628 0.6025874 0ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0.cp.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=0,
tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))## [1] "fitting model: Max.cor.Y.cv.0.cp.0.rpart"
## [1] " indep_vars: T.phase"
## Fitting cp = 0 on full training set
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1302
##
## CP nsplit rel error
## 1 0.3548951 0 1.0000000
## 2 0.0000000 1 0.6451049
##
## Variable importance
## T.phase
## 100
##
## Node number 1: 1302 observations, complexity param=0.3548951
## predicted class=N expected loss=0.4393241 P(node) =1
## class counts: 730 572
## probabilities: 0.561 0.439
## left son=2 (1005 obs) right son=3 (297 obs)
## Primary splits:
## T.phase < 0.5 to the left, improve=124.6249, (0 missing)
##
## Node number 2: 1005 observations
## predicted class=N expected loss=0.320398 P(node) =0.7718894
## class counts: 683 322
## probabilities: 0.680 0.320
##
## Node number 3: 297 observations
## predicted class=Y expected loss=0.1582492 P(node) =0.2281106
## class counts: 47 250
## probabilities: 0.158 0.842
##
## n= 1302
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1302 572 N (0.5606759 0.4393241)
## 2) T.phase< 0.5 1005 322 N (0.6796020 0.3203980) *
## 3) T.phase>=0.5 297 47 Y (0.1582492 0.8417508) *
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.6104589
## 3 0.2 0.6104589
## 4 0.3 0.6104589
## 5 0.4 0.5753740
## 6 0.5 0.5753740
## 7 0.6 0.5753740
## 8 0.7 0.5753740
## 9 0.8 0.5753740
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 N 730
## 2 Y 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 730
## 2 572
## Prediction
## Reference N Y
## N 0 730
## Y 0 572
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.393241e-01 0.000000e+00 4.121448e-01 4.667779e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.436635e-160
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6102117
## 3 0.2 0.6102117
## 4 0.3 0.6102117
## 5 0.4 0.5434174
## 6 0.5 0.5434174
## 7 0.6 0.5434174
## 8 0.7 0.5434174
## 9 0.8 0.5434174
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 N 313
## 2 Y 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 313
## 2 245
## Prediction
## Reference N Y
## N 0 313
## Y 0 245
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.390681e-01 0.000000e+00 3.974126e-01 4.813720e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.320342e-69
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.cv.0.cp.0.rpart rpart T.phase 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.473 0.017 0.6863397
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.3 0.6104589 0.4393241
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.4121448 0.4667779 0 0.6739975
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.6102117 0.4390681
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.3974126 0.481372 0if (glb_is_regression || glb_is_binomial) # For multinomials this model will be run next by default
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.rpart"
## [1] " indep_vars: T.phase"
## + Fold1: cp=0
## - Fold1: cp=0
## + Fold2: cp=0
## - Fold2: cp=0
## + Fold3: cp=0
## - Fold3: cp=0
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.177 on full training set
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1302
##
## CP nsplit rel error
## 1 0.3548951 0 1.0000000
## 2 0.0000000 1 0.6451049
##
## Variable importance
## T.phase
## 100
##
## Node number 1: 1302 observations, complexity param=0.3548951
## predicted class=N expected loss=0.4393241 P(node) =1
## class counts: 730 572
## probabilities: 0.561 0.439
## left son=2 (1005 obs) right son=3 (297 obs)
## Primary splits:
## T.phase < 0.5 to the left, improve=124.6249, (0 missing)
##
## Node number 2: 1005 observations
## predicted class=N expected loss=0.320398 P(node) =0.7718894
## class counts: 683 322
## probabilities: 0.680 0.320
##
## Node number 3: 297 observations
## predicted class=Y expected loss=0.1582492 P(node) =0.2281106
## class counts: 47 250
## probabilities: 0.158 0.842
##
## n= 1302
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1302 572 N (0.5606759 0.4393241)
## 2) T.phase< 0.5 1005 322 N (0.6796020 0.3203980) *
## 3) T.phase>=0.5 297 47 Y (0.1582492 0.8417508) *
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.6104589
## 3 0.2 0.6104589
## 4 0.3 0.6104589
## 5 0.4 0.5753740
## 6 0.5 0.5753740
## 7 0.6 0.5753740
## 8 0.7 0.5753740
## 9 0.8 0.5753740
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 N 730
## 2 Y 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 730
## 2 572
## Prediction
## Reference N Y
## N 0 730
## Y 0 572
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.393241e-01 0.000000e+00 4.121448e-01 4.667779e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.436635e-160
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6102117
## 3 0.2 0.6102117
## 4 0.3 0.6102117
## 5 0.4 0.5434174
## 6 0.5 0.5434174
## 7 0.6 0.5434174
## 8 0.7 0.5434174
## 9 0.8 0.5434174
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 N 313
## 2 Y 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.rpart.Y
## 1 313
## 2 245
## Prediction
## Reference N Y
## N 0 313
## Y 0 245
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.390681e-01 0.000000e+00 3.974126e-01 4.813720e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.320342e-69
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.rpart rpart T.phase 3
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.984 0.018 0.6863397
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.3 0.6104589 0.7165899
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.4121448 0.4667779 0.3930612 0.6739975
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.6102117 0.4390681
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.3974126 0.481372 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.02887088 0.06342212# Used to compare vs. Interactions.High.cor.Y
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.glm"
## [1] " indep_vars: T.phase"
## + Fold1: parameter=none
## - Fold1: parameter=none
## + Fold2: parameter=none
## - Fold2: parameter=none
## + Fold3: parameter=none
## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9202 -0.8789 -0.8789 0.5870 1.5088
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.7519 0.0676 -11.12 <2e-16 ***
## T.phase 2.4233 0.1728 14.03 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1785.7 on 1301 degrees of freedom
## Residual deviance: 1520.0 on 1300 degrees of freedom
## AIC: 1524
##
## Number of Fisher Scoring iterations: 4
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.6104589
## 3 0.2 0.6104589
## 4 0.3 0.6104589
## 5 0.4 0.5753740
## 6 0.5 0.5753740
## 7 0.6 0.5753740
## 8 0.7 0.5753740
## 9 0.8 0.5753740
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.Y
## 1 N 730
## 2 Y 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 730
## 2 572
## Prediction
## Reference N Y
## N 0 730
## Y 0 572
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.393241e-01 0.000000e+00 4.121448e-01 4.667779e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.436635e-160
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6102117
## 3 0.2 0.6102117
## 4 0.3 0.6102117
## 5 0.4 0.5434174
## 6 0.5 0.5434174
## 7 0.6 0.5434174
## 8 0.7 0.5434174
## 9 0.8 0.5434174
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.Y
## 1 N 313
## 2 Y 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Max.cor.Y.glm.Y
## 1 313
## 2 245
## Prediction
## Reference N Y
## N 0 313
## Y 0 245
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.390681e-01 0.000000e+00 3.974126e-01 4.813720e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.320342e-69
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.glm glm T.phase 1
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.873 0.021 0.6863397
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.3 0.6104589 0.7165899
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.4121448 0.4667779 0.3930612 0.6739975
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.6102117 0.4390681
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.3974126 0.481372 0 1524.042
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.02887088 0.06342212# Interactions.High.cor.Y
if (length(int_feats <- setdiff(unique(glb_feats_df$cor.high.X), NA)) > 0) {
# lm & glm handle interaction terms; rpart & rf do not
if (glb_is_regression || glb_is_binomial) {
indep_vars_vctr <-
c(max_cor_y_x_var, paste(max_cor_y_x_var, int_feats, sep=":"))
} else { indep_vars_vctr <- union(max_cor_y_x_var, int_feats) }
ret_lst <- myfit_mdl(model_id="Interact.High.cor.Y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)
} ## [1] "fitting model: Interact.High.cor.Y.glm"
## [1] " indep_vars: T.phase, T.phase:T.cancer, T.phase:A.nausea, T.phase:T.num.words, T.phase:T.num.words.unq, T.phase:A.point, T.phase:A.breast, T.phase:A.life"
## + Fold1: parameter=none
## - Fold1: parameter=none
## + Fold2: parameter=none
## - Fold2: parameter=none
## + Fold3: parameter=none
## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.4643 -0.8789 -0.8789 1.1710 1.5088
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.75194 0.06760 -11.123 < 2e-16 ***
## T.phase 2.57726 0.73909 3.487 0.000488 ***
## `T.phase:T.cancer` 1.43824 0.43772 3.286 0.001017 **
## `T.phase:A.nausea` 0.29650 0.42362 0.700 0.483974
## `T.phase:T.num.words` 0.54053 0.28108 1.923 0.054475 .
## `T.phase:T.num.words.unq` -0.64215 0.30314 -2.118 0.034151 *
## `T.phase:A.point` -0.19635 0.33926 -0.579 0.562752
## `T.phase:A.breast` -0.04041 0.14283 -0.283 0.777224
## `T.phase:A.life` -2.07589 0.63312 -3.279 0.001042 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1785.7 on 1301 degrees of freedom
## Residual deviance: 1486.9 on 1293 degrees of freedom
## AIC: 1504.9
##
## Number of Fisher Scoring iterations: 5
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 1 0
## Y 729 572
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 1
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 729
## 2 572
## Reference
## Prediction N Y
## N 1 0
## Y 729 572
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 1
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 729
## 2 572
## Reference
## Prediction N Y
## N 1 0
## Y 729 572
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 1
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 729
## 2 572
## Reference
## Prediction N Y
## N 688 322
## Y 42 250
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 688
## 2 Y 322
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 42
## 2 250
## Reference
## Prediction N Y
## N 690 326
## Y 40 246
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 690
## 2 Y 326
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 40
## 2 246
## Reference
## Prediction N Y
## N 693 332
## Y 37 240
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 693
## 2 Y 332
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 37
## 2 240
## Reference
## Prediction N Y
## N 698 339
## Y 32 233
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 698
## 2 Y 339
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 32
## 2 233
## Reference
## Prediction N Y
## N 704 358
## Y 26 214
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 704
## 2 Y 358
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 26
## 2 214
## Reference
## Prediction N Y
## N 725 476
## Y 5 96
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 725
## 2 Y 476
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 5
## 2 96
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.6107848
## 3 0.2 0.6107848
## 4 0.3 0.6107848
## 5 0.4 0.5787037
## 6 0.5 0.5734266
## 7 0.6 0.5653710
## 8 0.7 0.5567503
## 9 0.8 0.5270936
## 10 0.9 0.2852897
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 1
## 2 Y NA
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 729
## 2 572
## Reference
## Prediction N Y
## N 1 0
## Y 729 572
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 1
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 729
## 2 572
## Prediction
## Reference N Y
## N 1 729
## Y 0 572
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.400922e-01 1.203828e-03 4.129061e-01 4.675493e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 4.020083e-160
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 1 0
## Y 312 245
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 1
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 312
## 2 245
## Reference
## Prediction N Y
## N 2 0
## Y 311 245
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 2
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 311
## 2 245
## Reference
## Prediction N Y
## N 3 0
## Y 310 245
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 3
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 310
## 2 245
## Reference
## Prediction N Y
## N 301 148
## Y 12 97
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 301
## 2 Y 148
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 12
## 2 97
## Reference
## Prediction N Y
## N 302 150
## Y 11 95
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 302
## 2 Y 150
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 11
## 2 95
## Reference
## Prediction N Y
## N 303 151
## Y 10 94
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 303
## 2 Y 151
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 10
## 2 94
## Reference
## Prediction N Y
## N 304 160
## Y 9 85
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 304
## 2 Y 160
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 9
## 2 85
## Reference
## Prediction N Y
## N 308 164
## Y 5 81
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 308
## 2 Y 164
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 5
## 2 81
## Reference
## Prediction N Y
## N 312 209
## Y 1 36
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 312
## 2 Y 209
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 1
## 2 36
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6109726
## 3 0.2 0.6117353
## 4 0.3 0.6125000
## 5 0.4 0.5480226
## 6 0.5 0.5413105
## 7 0.6 0.5386819
## 8 0.7 0.5014749
## 9 0.8 0.4894260
## 10 0.9 0.2553191
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 3
## 2 Y NA
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 310
## 2 245
## Reference
## Prediction N Y
## N 3 0
## Y 310 245
## trial.fctr trial.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 3
## 2 Y 0
## trial.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 310
## 2 245
## Prediction
## Reference N Y
## N 3 310
## Y 0 245
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.444444e-01 8.426483e-03 4.027083e-01 4.867718e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 5.945837e-69
## model_id model_method
## 1 Interact.High.cor.Y.glm glm
## feats
## 1 T.phase, T.phase:T.cancer, T.phase:A.nausea, T.phase:T.num.words, T.phase:T.num.words.unq, T.phase:A.point, T.phase:A.breast, T.phase:A.life
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 0.949 0.046
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.6936201 0.3 0.6107848 0.7150538
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.4129061 0.4675493 0.3883401 0.6848667
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.6125 0.4444444
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.4027083 0.4867718 0.008426483 1504.884
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.02815715 0.06161482#Low.cor.X
if (glb_is_classification && glb_is_binomial)
indep_vars_vctr <- subset(glb_feats_df, is.na(cor.high.X) &
is.ConditionalX.y &
(exclude.as.feat != 1))[, "id"] else
indep_vars_vctr <- subset(glb_feats_df, is.na(cor.high.X) &
(exclude.as.feat != 1))[, "id"]
ret_lst <- myfit_mdl(model_id="Low.cor.X",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Low.cor.X.glm"
## [1] " indep_vars: T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, A.two, A.object, A.rang, A.paclitaxel, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use"
## + Fold1: parameter=none## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## - Fold1: parameter=none
## + Fold2: parameter=none## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## - Fold2: parameter=none
## + Fold3: parameter=none## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -8.49 0.00 0.00 0.00 8.49
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -7.212e+14 9.224e+06 -78190723 <2e-16 ***
## T.phase 1.389e+15 8.377e+06 165868151 <2e-16 ***
## A.toxic 2.110e+14 2.596e+06 81294833 <2e-16 ***
## A.mgm2 8.533e+13 1.950e+06 43767446 <2e-16 ***
## T.metastat 3.463e+14 9.037e+06 38314776 <2e-16 ***
## A.everi 3.760e+14 4.221e+06 89083278 <2e-16 ***
## A.median 1.053e+14 3.018e+06 34881522 <2e-16 ***
## A.rate 2.050e+14 2.530e+06 81022914 <2e-16 ***
## T.studi 2.595e+14 6.557e+06 39573640 <2e-16 ***
## T.versus 6.686e+14 7.298e+06 91613630 <2e-16 ***
## T.advanc 2.890e+14 8.555e+06 33779141 <2e-16 ***
## A.respons 8.092e+13 2.149e+06 37650290 <2e-16 ***
## A.metastat -3.407e+13 3.931e+06 -8667564 <2e-16 ***
## A.progress 5.195e+14 3.429e+06 151492939 <2e-16 ***
## A.advanc 2.899e+14 4.789e+06 60527223 <2e-16 ***
## A.neutropenia 3.606e+14 6.156e+06 58580331 <2e-16 ***
## A.partial 1.959e+13 6.518e+06 3005516 <2e-16 ***
## A.toler 3.208e+14 5.003e+06 64117606 <2e-16 ***
## A.combin 4.499e+13 2.484e+06 18111593 <2e-16 ***
## A.firstlin 6.474e+14 6.413e+06 100942162 <2e-16 ***
## A.regimen 1.938e+14 2.373e+06 81680102 <2e-16 ***
## A.cyclophosphamid 2.966e+14 4.579e+06 64770580 <2e-16 ***
## T.docetaxel 1.646e+14 1.126e+07 14619699 <2e-16 ***
## T.cyclophosphamid 6.278e+14 8.475e+06 74076232 <2e-16 ***
## A.overal 1.531e+14 3.412e+06 44872593 <2e-16 ***
## A.week -4.520e+13 1.878e+06 -24070284 <2e-16 ***
## A.month 1.359e+14 1.892e+06 71833473 <2e-16 ***
## A.surviv -1.923e+14 2.265e+06 -84924337 <2e-16 ***
## T.combin 7.145e+14 8.478e+06 84283779 <2e-16 ***
## A.day -6.571e+12 1.954e+06 -3362720 <2e-16 ***
## A.arm 1.015e+14 1.387e+06 73131518 <2e-16 ***
## A.docetaxel 3.409e+13 2.806e+06 12147676 <2e-16 ***
## T.cancer 5.013e+14 5.421e+06 92469983 <2e-16 ***
## T.iii -5.531e+14 1.052e+07 -52558833 <2e-16 ***
## T.compar 5.993e+14 8.969e+06 66820001 <2e-16 ***
## A.durat -4.189e+14 4.643e+06 -90219672 <2e-16 ***
## T.plus 1.588e+14 8.270e+06 19198569 <2e-16 ***
## A.given 3.542e+14 3.966e+06 89298225 <2e-16 ***
## A.epirubicin 1.397e+13 2.903e+06 4813360 <2e-16 ***
## A.mbc -6.654e+13 4.025e+06 -16531130 <2e-16 ***
## A.previous 3.232e+14 4.581e+06 70555150 <2e-16 ***
## A.X100 3.689e+14 4.472e+06 82483875 <2e-16 ***
## A.four 8.488e+13 4.206e+06 20182885 <2e-16 ***
## A.cycl 7.350e+13 2.139e+06 34363762 <2e-16 ***
## A.methotrex 2.063e+14 7.446e+06 27704818 <2e-16 ***
## T.group 3.971e+14 8.976e+06 44235492 <2e-16 ***
## A.phase 2.604e+14 4.773e+06 54544725 <2e-16 ***
## A.diseas -2.203e+13 3.098e+06 -7110055 <2e-16 ***
## A.efficaci 3.529e+13 3.917e+06 9009641 <2e-16 ***
## A.X500 4.033e+14 5.152e+06 78280222 <2e-16 ***
## A.grade -9.375e+13 3.289e+06 -28500027 <2e-16 ***
## A.patient 2.117e+13 1.126e+06 18791592 <2e-16 ***
## A.occur -2.366e+14 4.880e+06 -48481146 <2e-16 ***
## A.hundr 5.610e+14 6.963e+06 80564336 <2e-16 ***
## A.schedul -8.614e+12 4.094e+06 -2103951 <2e-16 ***
## T.doxorubicin -6.021e+14 1.091e+07 -55198241 <2e-16 ***
## A.either 4.443e+14 4.441e+06 100034303 <2e-16 ***
## A.dose -1.041e+14 1.871e+06 -55666479 <2e-16 ***
## A.activ 5.632e+13 3.417e+06 16480947 <2e-16 ***
## A.progressionfre 2.751e+13 7.723e+06 3561612 <2e-16 ***
## A.safeti 5.876e+14 5.927e+06 99139889 <2e-16 ***
## A.doxorubicin 7.327e+13 3.078e+06 23807015 <2e-16 ***
## A.seen 3.896e+14 5.619e+06 69335971 <2e-16 ***
## A.elig 2.783e+14 6.494e+06 42859996 <2e-16 ***
## T.trial -1.932e+14 6.181e+06 -31252913 <2e-16 ***
## A.hematolog -2.759e+14 7.806e+06 -35343248 <2e-16 ***
## A.plus -1.030e+14 2.900e+06 -35525205 <2e-16 ***
## A.respect -4.821e+13 3.281e+06 -14691811 <2e-16 ***
## A.fluorouracil 8.364e+13 7.779e+06 10752314 <2e-16 ***
## A.X5fluorouracil -3.317e+14 7.377e+06 -44966470 <2e-16 ***
## A.superior 2.029e+14 6.913e+06 29351269 <2e-16 ***
## A.enter -1.617e+14 9.207e+06 -17562154 <2e-16 ***
## A.two 2.130e+14 3.104e+06 68634219 <2e-16 ***
## A.object 1.520e+13 4.716e+06 3222784 <2e-16 ***
## A.rang 4.900e+14 5.464e+06 89676634 <2e-16 ***
## A.paclitaxel 5.576e+13 2.579e+06 21616406 <2e-16 ***
## A.three 5.240e+12 3.400e+06 1541362 <2e-16 ***
## T.random 1.879e+14 7.043e+06 26684390 <2e-16 ***
## A.prior 2.597e+14 4.947e+06 52498059 <2e-16 ***
## A.cmf 4.972e+13 2.204e+06 22563876 <2e-16 ***
## A.sequenti 9.705e+13 5.427e+06 17882470 <2e-16 ***
## A.oral 1.457e+14 3.047e+06 47832073 <2e-16 ***
## A.complet -2.189e+14 3.939e+06 -55580959 <2e-16 ***
## A.diseasefre 1.725e+14 5.228e+06 33003709 <2e-16 ***
## A.achiev 3.029e+13 5.220e+06 5802975 <2e-16 ***
## A.similar 1.169e+14 4.126e+06 28343159 <2e-16 ***
## T.randomis -1.500e+14 1.251e+07 -11995031 <2e-16 ***
## A.six -9.040e+13 4.641e+06 -19476665 <2e-16 ***
## A.secondari -3.986e+14 9.803e+06 -40656137 <2e-16 ***
## A.well 1.277e+13 5.986e+06 2133127 <2e-16 ***
## T.num.words.unq -7.784e+13 8.386e+05 -92815976 <2e-16 ***
## A.event 3.765e+14 4.466e+06 84304530 <2e-16 ***
## A.anthracyclin -1.409e+14 4.750e+06 -29668044 <2e-16 ***
## A.death 1.030e+14 6.467e+06 15921597 <2e-16 ***
## A.longer 6.209e+14 6.217e+06 99875205 <2e-16 ***
## A.eight 2.391e+14 7.622e+06 31370904 <2e-16 ***
## A.one -9.534e+13 3.827e+06 -24912176 <2e-16 ***
## A.enrol -8.608e+13 7.032e+06 -12240636 <2e-16 ***
## A.term 1.855e+14 7.025e+06 26402904 <2e-16 ***
## A.administ -1.426e+13 4.844e+06 -2942890 <2e-16 ***
## A.failur 4.088e+14 4.653e+06 87847897 <2e-16 ***
## A.common -1.115e+14 6.536e+06 -17063618 <2e-16 ***
## A.frequent -4.599e+13 7.391e+06 -6222179 <2e-16 ***
## A.assign 2.488e+14 4.665e+06 53328269 <2e-16 ***
## A.nodeposit 2.245e+14 6.762e+06 33192978 <2e-16 ***
## A.respond 2.365e+14 5.507e+06 42951156 <2e-16 ***
## A.endpoint 1.436e+14 6.767e+06 21215536 <2e-16 ***
## A.stabl -5.689e+14 7.853e+06 -72443025 <2e-16 ***
## A.confid 7.653e+14 8.644e+06 88534641 <2e-16 ***
## A.prolong 5.253e+14 8.355e+06 62875213 <2e-16 ***
## A.continu 6.248e+13 4.635e+06 13479909 <2e-16 ***
## A.daili 4.940e+14 3.944e+06 125261882 <2e-16 ***
## A.experienc 1.390e+14 6.567e+06 21169500 <2e-16 ***
## A.follow 1.619e+14 3.440e+06 47075343 <2e-16 ***
## A.treatment 5.100e+13 1.519e+06 33580040 <2e-16 ***
## A.time -2.233e+14 2.937e+06 -76030803 <2e-16 ***
## A.receiv -1.182e+13 1.967e+06 -6009441 <2e-16 ***
## A.dfs 3.060e+14 3.439e+06 88985876 <2e-16 ***
## A.primari 1.086e+14 3.429e+06 31673820 <2e-16 ***
## A.interv -1.524e+14 6.000e+06 -25395452 <2e-16 ***
## A.compar 1.355e+14 2.571e+06 52709709 <2e-16 ***
## A.total 5.265e+14 4.535e+06 116094381 <2e-16 ***
## T.treatment 1.352e+14 6.099e+06 22162740 <2e-16 ***
## A.intraven -5.622e+14 6.589e+06 -85317148 <2e-16 ***
## T.tamoxifen 5.069e+14 7.567e+06 66994090 <2e-16 ***
## T.result -6.615e+13 7.744e+06 -8542901 <2e-16 ***
## A.cours -1.388e+14 4.014e+06 -34582138 <2e-16 ***
## A.observ -1.944e+13 3.112e+06 -6247403 <2e-16 ***
## A.advers -3.627e+14 6.091e+06 -59541737 <2e-16 ***
## A.evalu -1.599e+14 3.082e+06 -51879070 <2e-16 ***
## A.versus 1.578e+14 2.867e+06 55033492 <2e-16 ***
## A.pretreat -1.400e+14 5.465e+06 -25613756 <2e-16 ***
## A.point 2.253e+14 6.370e+06 35374837 <2e-16 ***
## T.postmenopaus 2.846e+14 1.050e+07 27115897 <2e-16 ***
## A.includ -2.432e+14 4.004e+06 -60742763 <2e-16 ***
## A.seven -4.827e+14 9.142e+06 -52794428 <2e-16 ***
## A.nausea -2.793e+13 4.628e+06 -6036483 <2e-16 ***
## .rnorm 3.809e+13 2.204e+06 17276788 <2e-16 ***
## A.five 1.855e+14 7.083e+06 26189841 <2e-16 ***
## A.better -2.278e+14 6.657e+06 -34225676 <2e-16 ***
## A.multicent -6.699e+14 9.267e+06 -72292925 <2e-16 ***
## A.oper 2.695e+14 6.825e+06 39482183 <2e-16 ***
## A.stage 7.168e+14 4.803e+06 149248993 <2e-16 ***
## A.alon -3.981e+13 3.282e+06 -12129391 <2e-16 ***
## A.less -1.830e+14 4.112e+06 -44506194 <2e-16 ***
## A.random -1.739e+13 3.291e+06 -5285448 <2e-16 ***
## A.although -1.152e+14 7.550e+06 -15260858 <2e-16 ***
## A.addit 1.727e+14 4.041e+06 42739691 <2e-16 ***
## A.agent -4.868e+13 5.129e+06 -9491368 <2e-16 ***
## A.due -1.657e+14 6.760e+06 -24511510 <2e-16 ***
## A.andor 5.564e+14 7.582e+06 73381072 <2e-16 ***
## A.lymph 2.852e+14 5.677e+06 50242719 <2e-16 ***
## A.followup 5.015e+13 4.723e+06 10618918 <2e-16 ***
## A.postmenopaus -6.801e+13 4.206e+06 -16171782 <2e-16 ***
## A.involv 2.530e+14 6.446e+06 39252777 <2e-16 ***
## A.side 2.681e+14 6.856e+06 39094998 <2e-16 ***
## A.standard 4.421e+14 4.203e+06 105195487 <2e-16 ***
## A.receptorposit 1.824e+14 7.104e+06 25680439 <2e-16 ***
## A.conduct -2.930e+14 7.725e+06 -37934160 <2e-16 ***
## A.neoadjuv 1.527e+14 4.227e+06 36112393 <2e-16 ***
## A.major 6.009e+14 9.165e+06 65568234 <2e-16 ***
## A.administr 3.239e+14 4.890e+06 66246853 <2e-16 ***
## A.limit -1.980e+14 7.500e+06 -26402229 <2e-16 ***
## A.consist 2.110e+13 6.919e+06 3050152 <2e-16 ***
## A.differ -2.003e+13 2.811e+06 -7127144 <2e-16 ***
## A.num.words.unq 7.007e+12 2.765e+05 25339741 <2e-16 ***
## A.singl 2.935e+14 7.355e+06 39907250 <2e-16 ***
## A.hazard -1.437e+14 6.893e+06 -20846550 <2e-16 ***
## A.sever -7.550e+13 4.752e+06 -15889612 <2e-16 ***
## A.result 3.065e+14 3.380e+06 90684974 <2e-16 ***
## A.randomis -2.288e+14 5.763e+06 -39695842 <2e-16 ***
## A.appear 2.940e+14 6.940e+06 42364129 <2e-16 ***
## A.treat -6.387e+13 2.872e+06 -22238382 <2e-16 ***
## A.profil -6.747e+14 6.471e+06 -104258791 <2e-16 ***
## A.year 1.408e+14 2.265e+06 62151140 <2e-16 ***
## A.accord 2.498e+14 6.787e+06 36801248 <2e-16 ***
## T.patient 2.172e+13 5.338e+06 4068949 <2e-16 ***
## A.site 1.554e+10 6.285e+06 2472 <2e-16 ***
## A.improv 7.891e+13 3.466e+06 22763011 <2e-16 ***
## A.tamoxifen 1.054e+14 1.660e+06 63505836 <2e-16 ***
## A.ratio -1.646e+14 6.311e+06 -26081654 <2e-16 ***
## A.evid 3.570e+14 7.509e+06 47551354 <2e-16 ***
## A.iii -1.173e+14 6.380e+06 -18391368 <2e-16 ***
## A.conclus 1.313e+14 7.020e+06 18711286 <2e-16 ***
## A.demonstr 4.321e+13 5.548e+06 7787774 <2e-16 ***
## A.statist 2.132e+14 4.430e+06 48114897 <2e-16 ***
## A.among 1.996e+14 4.110e+06 48579123 <2e-16 ***
## A.relaps -1.857e+14 3.926e+06 -47285757 <2e-16 ***
## A.human 9.628e+13 9.275e+06 10380503 <2e-16 ***
## A.without -1.756e+14 4.796e+06 -36610376 <2e-16 ***
## A.surgeri 2.025e+14 4.010e+06 50500880 <2e-16 ***
## A.drug 1.411e+14 4.057e+06 34782737 <2e-16 ***
## A.infus -1.581e+14 4.583e+06 -34504576 <2e-16 ***
## T.therapi -7.619e+13 6.944e+06 -10971272 <2e-16 ***
## A.higher -1.273e+14 4.308e+06 -29546274 <2e-16 ***
## A.purpos -1.307e+15 6.866e+06 -190350621 <2e-16 ***
## A.premenopaus 1.726e+14 4.158e+06 41523355 <2e-16 ***
## A.num.chars 4.031e+11 3.384e+04 11912360 <2e-16 ***
## A.background -7.190e+14 7.243e+06 -99257201 <2e-16 ***
## A.benefit -3.442e+14 3.905e+06 -88150196 <2e-16 ***
## A.growth 6.281e+12 6.879e+06 913086 <2e-16 ***
## T.chemotherapi 4.732e+14 6.521e+06 72564783 <2e-16 ***
## A.confirm 2.839e+14 7.136e+06 39791297 <2e-16 ***
## A.inhibitor 4.203e+14 1.015e+07 41423335 <2e-16 ***
## A.characterist -3.092e+14 8.682e+06 -35609251 <2e-16 ***
## A.whether -6.271e+13 6.436e+06 -9742855 <2e-16 ***
## A.popul -1.105e+13 5.852e+06 -1888992 <2e-16 ***
## A.initi -2.203e+14 5.163e+06 -42668840 <2e-16 ***
## A.group 2.897e+13 1.164e+06 24882184 <2e-16 ***
## A.mastectomi 5.994e+13 6.555e+06 9143522 <2e-16 ***
## A.least -4.220e+14 6.927e+06 -60910140 <2e-16 ***
## A.consid 4.489e+14 7.801e+06 57543819 <2e-16 ***
## A.node -2.648e+14 4.606e+06 -57487160 <2e-16 ***
## A.axillari -6.000e+13 3.884e+06 -15446327 <2e-16 ***
## A.num.words -2.322e+13 4.287e+05 -54160345 <2e-16 ***
## A.caus -4.587e+14 8.628e+06 -53167032 <2e-16 ***
## A.endocrin -2.108e+14 3.774e+06 -55850496 <2e-16 ***
## A.estim -1.314e+14 6.982e+06 -18817998 <2e-16 ***
## A.new -3.164e+14 5.751e+06 -55020083 <2e-16 ***
## A.postop 1.887e+14 5.061e+06 37278810 <2e-16 ***
## A.distant 4.032e+14 7.338e+06 54951358 <2e-16 ***
## A.regard 2.279e+14 8.766e+06 26002760 <2e-16 ***
## A.requir 7.131e+13 6.835e+06 10432255 <2e-16 ***
## A.defin 7.655e+14 8.829e+06 86700084 <2e-16 ***
## A.trend 2.188e+14 7.855e+06 27858547 <2e-16 ***
## A.subgroup -1.995e+14 5.771e+06 -34573611 <2e-16 ***
## A.receptor 2.373e+14 4.732e+06 50157843 <2e-16 ***
## A.greater 9.007e+13 4.933e+06 18259021 <2e-16 ***
## A.therapi -9.182e+12 1.996e+06 -4599354 <2e-16 ***
## A.studi 6.034e+13 2.122e+06 28443040 <2e-16 ***
## A.howev 5.772e+14 5.638e+06 102367593 <2e-16 ***
## A.support -1.912e+14 5.039e+06 -37936811 <2e-16 ***
## T.adjuv 4.888e+14 6.894e+06 70892768 <2e-16 ***
## A.number 1.481e+14 4.559e+06 32481953 <2e-16 ***
## A.develop 9.959e+13 5.028e+06 19808201 <2e-16 ***
## A.comparison -3.249e+14 8.231e+06 -39468189 <2e-16 ***
## A.prospect 2.402e+14 6.354e+06 37801169 <2e-16 ***
## A.regress 2.578e+14 8.406e+06 30668457 <2e-16 ***
## A.per -1.615e+13 4.919e+06 -3282180 <2e-16 ***
## A.age -6.141e+13 3.736e+06 -16438371 <2e-16 ***
## A.general -5.601e+14 7.490e+06 -74770714 <2e-16 ***
## A.earli -1.455e+14 4.239e+06 -34319506 <2e-16 ***
## A.X001 2.543e+14 7.006e+06 36297199 <2e-16 ***
## A.base -2.747e+14 8.399e+06 -32707990 <2e-16 ***
## A.carcinoma -8.055e+13 4.546e+06 -17718116 <2e-16 ***
## A.method 1.018e+14 5.563e+06 18301406 <2e-16 ***
## A.patholog 1.255e+13 4.461e+06 2813338 <2e-16 ***
## A.set 5.407e+12 6.004e+06 900501 <2e-16 ***
## A.her2 4.158e+13 2.707e+06 15361050 <2e-16 ***
## A.status -2.665e+14 3.599e+06 -74044927 <2e-16 ***
## A.start -5.653e+13 7.445e+06 -7593255 <2e-16 ***
## A.incid -1.361e+14 4.445e+06 -30612493 <2e-16 ***
## A.determin 1.844e+13 5.771e+06 3196117 <2e-16 ***
## A.posit 1.920e+14 3.684e+06 52118920 <2e-16 ***
## A.progesteron 2.293e+14 9.615e+06 23850754 <2e-16 ***
## A.aromatas -5.046e+14 1.080e+07 -46744510 <2e-16 ***
## A.local -4.378e+14 4.545e+06 -96322676 <2e-16 ***
## A.main 2.559e+13 7.922e+06 3230703 <2e-16 ***
## A.present -7.140e+14 5.764e+06 -123874295 <2e-16 ***
## A.aim 3.003e+13 7.055e+06 4257088 <2e-16 ***
## A.possibl -2.502e+14 7.420e+06 -33721211 <2e-16 ***
## A.perform -5.935e+13 4.953e+06 -11983228 <2e-16 ***
## A.larg -1.473e+13 8.656e+06 -1701753 <2e-16 ***
## A.shown -2.535e+14 8.418e+06 -30109473 <2e-16 ***
## A.hormon -6.359e+13 3.873e+06 -16417867 <2e-16 ***
## A.period 3.348e+13 7.174e+06 4666977 <2e-16 ***
## A.reduct 2.807e+14 4.426e+06 63412171 <2e-16 ***
## T.clinic 3.978e+14 9.023e+06 44092351 <2e-16 ***
## A.second 6.046e+13 7.708e+06 7843896 <2e-16 ***
## A.breast -2.024e+13 1.910e+06 -10598346 <2e-16 ***
## A.estrogen -4.399e+14 4.258e+06 -103298571 <2e-16 ***
## A.tissu 1.524e+14 4.396e+06 34679842 <2e-16 ***
## A.adjuv 7.092e+13 2.593e+06 27354074 <2e-16 ***
## A.case 1.038e+14 4.065e+06 25536841 <2e-16 ***
## A.find 5.769e+14 7.065e+06 81655472 <2e-16 ***
## A.obtain -4.008e+14 7.802e+06 -51366814 <2e-16 ***
## A.suggest 3.221e+14 6.134e+06 52509774 <2e-16 ***
## A.within -9.894e+13 6.654e+06 -14869934 <2e-16 ***
## A.remain -1.791e+14 7.240e+06 -24737643 <2e-16 ***
## A.design -3.262e+13 6.719e+06 -4854826 <2e-16 ***
## A.X0001 4.483e+13 4.253e+06 10541338 <2e-16 ***
## A.potenti -1.486e+14 7.942e+06 -18713365 <2e-16 ***
## A.import -2.258e+13 8.727e+06 -2587166 <2e-16 ***
## A.detect 5.851e+13 6.494e+06 9010273 <2e-16 ***
## A.women 5.526e+13 2.155e+06 25639052 <2e-16 ***
## A.recurr -6.313e+13 3.495e+06 -18062286 <2e-16 ***
## A.investig -5.284e+14 5.373e+06 -98345465 <2e-16 ***
## A.chemotherapi -1.416e+14 1.623e+06 -87242422 <2e-16 ***
## A.histolog -4.909e+14 7.788e+06 -63040396 <2e-16 ***
## A.lower 1.436e+14 6.689e+06 21465775 <2e-16 ***
## T.earli -1.769e+14 1.002e+07 -17653032 <2e-16 ***
## A.select 1.502e+14 8.145e+06 18443419 <2e-16 ***
## A.first -1.913e+14 3.865e+06 -49478272 <2e-16 ***
## A.signific -1.705e+14 2.250e+06 -75774090 <2e-16 ***
## A.low 8.565e+13 5.954e+06 14385995 <2e-16 ***
## A.mean -1.621e+14 4.396e+06 -36870117 <2e-16 ***
## T.women -6.202e+14 8.451e+06 -73381667 <2e-16 ***
## A.radiotherapi 3.032e+13 3.456e+06 8775074 <2e-16 ***
## A.clinic 1.783e+14 3.017e+06 59105065 <2e-16 ***
## A.also -2.510e+14 4.839e+06 -51872712 <2e-16 ***
## A.relat -1.474e+14 4.654e+06 -31667206 <2e-16 ***
## A.life 3.810e+14 5.394e+06 70636804 <2e-16 ***
## A.size -5.032e+13 5.555e+06 -9059143 <2e-16 ***
## A.effect 9.986e+13 2.657e+06 37576819 <2e-16 ***
## A.outcom -1.085e+14 4.980e+06 -21790546 <2e-16 ***
## A.wherea -3.304e+14 6.906e+06 -47835432 <2e-16 ***
## A.may -3.721e+13 4.557e+06 -8164719 <2e-16 ***
## A.indic 2.194e+14 5.650e+06 38836595 <2e-16 ***
## A.system 5.300e+12 5.112e+06 1036776 <2e-16 ***
## A.test 9.377e+13 4.954e+06 18928126 <2e-16 ***
## A.metastas 1.617e+14 3.631e+06 44532860 <2e-16 ***
## A.type 2.230e+14 7.317e+06 30481200 <2e-16 ***
## A.function. 4.500e+10 5.325e+06 8450 <2e-16 ***
## A.analysi -2.435e+13 4.018e+06 -6059910 <2e-16 ***
## A.avail 2.534e+14 9.353e+06 27092150 <2e-16 ***
## A.negat -3.269e+14 4.959e+06 -65913874 <2e-16 ***
## A.analys -5.618e+13 6.670e+06 -8422208 <2e-16 ***
## A.X005 -4.441e+14 7.158e+06 -62050693 <2e-16 ***
## A.report -4.958e+14 4.576e+06 -108343909 <2e-16 ***
## A.show 3.752e+13 3.837e+06 9779363 <2e-16 ***
## A.provid -2.742e+14 7.710e+06 -35568731 <2e-16 ***
## A.tumor 1.632e+14 2.064e+06 79078903 <2e-16 ***
## A.can -2.022e+14 6.789e+06 -29787036 <2e-16 ***
## A.doubleblind -1.003e+15 9.403e+06 -106718502 <2e-16 ***
## A.tumour -9.455e+13 2.958e+06 -31967810 <2e-16 ***
## A.serum 3.446e+13 2.783e+06 12380920 <2e-16 ***
## A.factor -7.436e+12 4.092e+06 -1817093 <2e-16 ***
## A.analyz 7.516e+13 8.079e+06 9303074 <2e-16 ***
## A.proport 2.901e+14 7.998e+06 36269388 <2e-16 ***
## A.valu -2.296e+14 4.964e+06 -46262545 <2e-16 ***
## A.sampl -4.050e+13 5.615e+06 -7211957 <2e-16 ***
## A.inform 3.253e+12 9.804e+06 331806 <2e-16 ***
## A.assess 1.305e+14 3.129e+06 41719294 <2e-16 ***
## A.need -4.558e+14 8.183e+06 -55705020 <2e-16 ***
## A.independ 1.827e+13 7.284e+06 2508638 <2e-16 ***
## A.correl -1.574e+14 5.312e+06 -29628050 <2e-16 ***
## A.increas 6.310e+13 2.883e+06 21882233 <2e-16 ***
## A.level -1.016e+13 2.209e+06 -4598694 <2e-16 ***
## A.placebo 1.265e+14 2.865e+06 44152640 <2e-16 ***
## A.multivari -8.280e+14 9.418e+06 -87917491 <2e-16 ***
## A.model 2.391e+14 6.816e+06 35081794 <2e-16 ***
## A.trial -1.301e+14 2.396e+06 -54316895 <2e-16 ***
## A.high -1.863e+14 4.269e+06 -43645710 <2e-16 ***
## A.score -2.187e+14 3.954e+06 -55318253 <2e-16 ***
## A.express -8.425e+13 3.053e+06 -27598729 <2e-16 ***
## A.found -5.979e+13 5.538e+06 -10796939 <2e-16 ***
## A.particip -1.213e+14 6.918e+06 -17533404 <2e-16 ***
## A.control -1.912e+14 2.702e+06 -70783615 <2e-16 ***
## A.examin -2.907e+14 6.328e+06 -45931278 <2e-16 ***
## A.prognost 1.110e+14 5.742e+06 19334122 <2e-16 ***
## A.marker -2.013e+14 4.746e+06 -42406598 <2e-16 ***
## A.reduc -4.191e+14 4.304e+06 -97390811 <2e-16 ***
## A.cell -1.566e+14 3.922e+06 -39921804 <2e-16 ***
## A.measur -4.566e+13 3.726e+06 -12253984 <2e-16 ***
## A.chang -5.587e+13 3.685e+06 -15162264 <2e-16 ***
## T.respons -6.257e+14 1.118e+07 -55979444 <2e-16 ***
## A.baselin -9.590e+12 4.455e+06 -2152694 <2e-16 ***
## A.identifi -3.206e+14 7.243e+06 -44272184 <2e-16 ***
## A.decreas 3.147e+14 4.215e+06 74649336 <2e-16 ***
## A.bone -1.518e+14 2.506e+06 -60582165 <2e-16 ***
## A.prevent -2.080e+14 6.400e+06 -32500683 <2e-16 ***
## A.data -1.103e+14 4.290e+06 -25721367 <2e-16 ***
## A.associ 2.456e+14 3.469e+06 70784790 <2e-16 ***
## A.predict -2.905e+13 4.078e+06 -7123028 <2e-16 ***
## T.effect -5.174e+14 7.673e+06 -67434208 <2e-16 ***
## A.risk -1.148e+14 2.810e+06 -40859084 <2e-16 ***
## A.use -9.807e+13 2.598e+06 -37740355 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1785.7 on 1301 degrees of freedom
## Residual deviance: 5622.8 on 935 degrees of freedom
## AIC: 6356.8
##
## Number of Fisher Scoring iterations: 23
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.9307282
## 3 0.2 0.9307282
## 4 0.3 0.9307282
## 5 0.4 0.9307282
## 6 0.5 0.9307282
## 7 0.6 0.9307282
## 8 0.7 0.9307282
## 9 0.8 0.9307282
## 10 0.9 0.9307282
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.9000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Reference
## Prediction N Y
## N 700 48
## Y 30 524
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 700
## 2 Y 48
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 30
## 2 524
## Prediction
## Reference N Y
## N 700 30
## Y 48 524
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.400922e-01 8.779780e-01 9.257933e-01 9.523607e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 2.889418e-209 5.424550e-02
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6832298
## 3 0.2 0.6832298
## 4 0.3 0.6832298
## 5 0.4 0.6832298
## 6 0.5 0.6832298
## 7 0.6 0.6832298
## 8 0.7 0.6832298
## 9 0.8 0.6832298
## 10 0.9 0.6832298
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.9000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Reference
## Prediction N Y
## N 240 80
## Y 73 165
## trial.fctr trial.fctr.predict.Low.cor.X.glm.N
## 1 N 240
## 2 Y 80
## trial.fctr.predict.Low.cor.X.glm.Y
## 1 73
## 2 165
## Prediction
## Reference N Y
## N 240 73
## Y 80 165
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.258065e-01 4.416131e-01 6.867647e-01 7.624378e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 6.616923e-16 6.276258e-01
## model_id model_method
## 1 Low.cor.X.glm glm
## feats
## 1 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, A.two, A.object, A.rang, A.paclitaxel, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 18.582 5.48
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.937494 0.9 0.9307282 0.7058372
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9257933 0.9523607 0.4082901 0.7201213
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.9 0.6832298 0.7258065
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.6867647 0.7624378 0.4416131 6356.81
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.0146333 0.03093465glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.models",
chunk_step_major=glb_script_df[nrow(glb_script_df), "chunk_step_major"],
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed8 fit.models 5 0 40.403
## elapsed9 fit.models 5 1 89.153# All X that is not user excluded
if (glb_is_classification && glb_is_binomial) {
model_id_pfx <- "Conditional.X"
# indep_vars_vctr <- setdiff(names(glb_trnent_df), union(glb_rsp_var, glb_exclude_vars_as_features))
indep_vars_vctr <- subset(glb_feats_df, is.ConditionalX.y &
(exclude.as.feat != 1))[, "id"]
} else {
model_id_pfx <- "All.X"
indep_vars_vctr <- subset(glb_feats_df,
(exclude.as.feat != 1))[, "id"]
}
for (method in glb_models_method_vctr) {
ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ""), model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
# Since caret does not optimize rpart well
if (method == "rpart")
ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ".cp.0"), model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=0, tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))
# Compare how rf performs w/i & w/o .rnorm
if (method == "rf")
ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ".no.rnorm"), model_method=method,
indep_vars_vctr=setdiff(indep_vars_vctr, c(".rnorm")),
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
}## [1] "fitting model: Conditional.X.glm"
## [1] " indep_vars: T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use"
## + Fold1: parameter=none## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## - Fold1: parameter=none
## + Fold2: parameter=none## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## - Fold2: parameter=none
## + Fold3: parameter=none## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -8.49 0.00 0.00 0.00 8.49
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -6.206e+14 9.531e+06 -65111696 <2e-16 ***
## T.phase 1.015e+15 8.402e+06 120849854 <2e-16 ***
## A.toxic 2.254e+14 2.601e+06 86660465 <2e-16 ***
## A.mgm2 7.267e+13 1.960e+06 37071971 <2e-16 ***
## T.metastat 2.679e+14 9.144e+06 29295318 <2e-16 ***
## A.everi 2.182e+14 4.254e+06 51293841 <2e-16 ***
## A.median 5.613e+13 3.029e+06 18533252 <2e-16 ***
## A.rate 1.718e+14 2.549e+06 67376907 <2e-16 ***
## T.studi 5.318e+14 6.609e+06 80461257 <2e-16 ***
## T.versus 7.694e+14 7.627e+06 100877026 <2e-16 ***
## T.advanc 4.019e+14 8.740e+06 45984878 <2e-16 ***
## A.respons 6.848e+13 2.156e+06 31765790 <2e-16 ***
## A.metastat -1.241e+14 3.965e+06 -31286970 <2e-16 ***
## A.progress 3.524e+14 3.468e+06 101618434 <2e-16 ***
## A.advanc 6.930e+13 4.840e+06 14318395 <2e-16 ***
## A.neutropenia 4.849e+14 6.162e+06 78702954 <2e-16 ***
## A.partial -1.380e+13 6.542e+06 -2109422 <2e-16 ***
## A.toler 3.674e+14 5.033e+06 73003644 <2e-16 ***
## A.combin 7.865e+13 2.500e+06 31452170 <2e-16 ***
## A.firstlin 6.462e+14 6.420e+06 100651606 <2e-16 ***
## A.regimen 2.343e+14 2.382e+06 98354006 <2e-16 ***
## T.breast 5.147e+14 8.602e+06 59831919 <2e-16 ***
## A.cyclophosphamid 2.361e+14 4.600e+06 51329380 <2e-16 ***
## T.docetaxel 9.038e+12 1.130e+07 799701 <2e-16 ***
## T.cyclophosphamid 7.000e+14 9.142e+06 76568775 <2e-16 ***
## A.overal 9.351e+13 3.421e+06 27332530 <2e-16 ***
## A.week -7.224e+13 1.887e+06 -38277536 <2e-16 ***
## A.month 1.016e+14 1.898e+06 53510108 <2e-16 ***
## A.surviv -1.494e+14 2.274e+06 -65728063 <2e-16 ***
## T.combin 3.294e+14 8.571e+06 38427539 <2e-16 ***
## A.day -3.793e+13 1.964e+06 -19314427 <2e-16 ***
## A.arm 8.241e+13 1.394e+06 59123101 <2e-16 ***
## A.docetaxel 8.784e+13 2.819e+06 31159975 <2e-16 ***
## T.cancer 1.157e+13 7.353e+06 1573521 <2e-16 ***
## T.iii -2.249e+14 1.077e+07 -20879838 <2e-16 ***
## T.compar 5.094e+14 8.995e+06 56630227 <2e-16 ***
## A.durat -2.233e+14 4.652e+06 -47991327 <2e-16 ***
## T.plus 1.733e+14 8.819e+06 19646988 <2e-16 ***
## A.given 2.907e+14 3.979e+06 73066873 <2e-16 ***
## A.epirubicin 2.081e+13 2.914e+06 7142064 <2e-16 ***
## A.mbc -2.251e+14 4.041e+06 -55714754 <2e-16 ***
## A.previous 3.398e+14 4.594e+06 73972216 <2e-16 ***
## A.X100 4.353e+14 4.505e+06 96621900 <2e-16 ***
## A.four 1.654e+14 4.223e+06 39172450 <2e-16 ***
## A.cycl 5.251e+12 2.144e+06 2449102 <2e-16 ***
## A.methotrex 2.427e+14 7.462e+06 32522180 <2e-16 ***
## T.group 2.931e+14 9.571e+06 30628485 <2e-16 ***
## A.phase 1.579e+14 4.779e+06 33045530 <2e-16 ***
## A.diseas 5.212e+13 3.112e+06 16749170 <2e-16 ***
## A.efficaci 2.621e+13 3.930e+06 6670262 <2e-16 ***
## A.X500 3.457e+14 5.168e+06 66899918 <2e-16 ***
## A.grade -5.600e+13 3.302e+06 -16961286 <2e-16 ***
## A.patient -3.189e+13 1.132e+06 -28164171 <2e-16 ***
## A.occur -2.289e+14 4.901e+06 -46709828 <2e-16 ***
## A.hundr 3.836e+14 6.986e+06 54911105 <2e-16 ***
## A.schedul 1.087e+14 4.114e+06 26432532 <2e-16 ***
## T.doxorubicin -1.590e+14 1.092e+07 -14556375 <2e-16 ***
## A.either 3.933e+14 4.462e+06 88147651 <2e-16 ***
## A.dose -7.111e+13 1.887e+06 -37679059 <2e-16 ***
## A.activ 5.230e+13 3.441e+06 15199283 <2e-16 ***
## A.progressionfre 4.436e+13 7.772e+06 5706793 <2e-16 ***
## A.safeti 4.826e+14 5.954e+06 81058470 <2e-16 ***
## A.doxorubicin 5.529e+13 3.085e+06 17920954 <2e-16 ***
## A.seen 2.522e+14 5.641e+06 44702835 <2e-16 ***
## A.elig 4.238e+13 6.508e+06 6511671 <2e-16 ***
## T.trial -8.771e+13 6.259e+06 -14011679 <2e-16 ***
## A.hematolog -2.388e+14 7.830e+06 -30498594 <2e-16 ***
## A.plus -5.021e+13 2.907e+06 -17270169 <2e-16 ***
## A.vomit 4.357e+14 9.352e+06 46594863 <2e-16 ***
## A.respect -8.963e+13 3.285e+06 -27287879 <2e-16 ***
## A.fluorouracil 2.084e+14 7.799e+06 26719385 <2e-16 ***
## A.X5fluorouracil -2.421e+14 7.395e+06 -32735548 <2e-16 ***
## A.superior 2.799e+14 6.937e+06 40349475 <2e-16 ***
## A.enter 4.355e+14 9.258e+06 47039620 <2e-16 ***
## T.num.chars -4.342e+12 2.007e+05 -21633059 <2e-16 ***
## A.two 1.159e+14 3.108e+06 37297200 <2e-16 ***
## A.object 5.480e+13 4.754e+06 11527647 <2e-16 ***
## A.rang 4.384e+14 5.490e+06 79846239 <2e-16 ***
## A.paclitaxel 7.411e+13 2.592e+06 28595872 <2e-16 ***
## T.num.words -1.682e+14 4.193e+06 -40121557 <2e-16 ***
## A.three 5.059e+13 3.421e+06 14788970 <2e-16 ***
## T.random 1.847e+14 7.090e+06 26051136 <2e-16 ***
## A.prior 1.304e+14 5.011e+06 26021340 <2e-16 ***
## A.cmf 2.501e+13 2.210e+06 11319065 <2e-16 ***
## A.sequenti -1.877e+13 5.453e+06 -3441182 <2e-16 ***
## A.oral 4.788e+13 3.064e+06 15624421 <2e-16 ***
## A.complet -1.806e+14 3.953e+06 -45697485 <2e-16 ***
## A.diseasefre 1.887e+14 5.237e+06 36033738 <2e-16 ***
## A.achiev 7.484e+13 5.236e+06 14293220 <2e-16 ***
## A.similar -9.341e+13 4.153e+06 -22494031 <2e-16 ***
## T.randomis 1.922e+14 1.253e+07 15332828 <2e-16 ***
## A.six 5.228e+13 4.644e+06 11256951 <2e-16 ***
## A.secondari -4.769e+14 1.002e+07 -47604039 <2e-16 ***
## A.well -3.035e+13 6.039e+06 -5025460 <2e-16 ***
## T.num.words.unq 1.412e+14 3.594e+06 39300883 <2e-16 ***
## A.event 4.608e+14 4.483e+06 102800283 <2e-16 ***
## A.anthracyclin -2.210e+14 4.779e+06 -46239399 <2e-16 ***
## A.death -1.051e+13 6.475e+06 -1622753 <2e-16 ***
## A.longer 5.758e+14 6.236e+06 92322519 <2e-16 ***
## A.eight -4.596e+12 7.646e+06 -601125 <2e-16 ***
## A.one -3.199e+13 3.839e+06 -8333266 <2e-16 ***
## A.enrol -1.154e+14 7.063e+06 -16343968 <2e-16 ***
## A.term 3.414e+14 7.054e+06 48398746 <2e-16 ***
## A.administ 1.158e+14 4.862e+06 23824735 <2e-16 ***
## A.failur 2.308e+14 4.668e+06 49433068 <2e-16 ***
## A.common 2.228e+13 6.555e+06 3398695 <2e-16 ***
## A.frequent 2.262e+14 7.411e+06 30515277 <2e-16 ***
## A.assign 2.001e+14 4.688e+06 42681979 <2e-16 ***
## A.nodeposit 4.744e+13 6.796e+06 6979962 <2e-16 ***
## A.respond 2.665e+14 5.527e+06 48214989 <2e-16 ***
## A.endpoint 2.680e+14 6.872e+06 38995230 <2e-16 ***
## A.stabl -3.852e+14 7.887e+06 -48836085 <2e-16 ***
## A.confid 4.918e+14 8.656e+06 56816483 <2e-16 ***
## A.prolong 2.608e+14 8.367e+06 31170838 <2e-16 ***
## A.continu 3.403e+13 4.667e+06 7291265 <2e-16 ***
## A.daili 2.942e+14 3.954e+06 74407279 <2e-16 ***
## A.experienc 1.182e+13 6.594e+06 1793240 <2e-16 ***
## A.follow 1.009e+14 3.452e+06 29218282 <2e-16 ***
## A.treatment 9.711e+12 1.521e+06 6382761 <2e-16 ***
## A.end 6.876e+14 8.586e+06 80083793 <2e-16 ***
## A.time -1.157e+14 3.008e+06 -38467988 <2e-16 ***
## A.receiv 6.862e+13 1.972e+06 34794661 <2e-16 ***
## A.dfs 1.994e+14 3.442e+06 57916150 <2e-16 ***
## A.primari 6.250e+13 3.460e+06 18062271 <2e-16 ***
## A.interv -1.106e+14 6.017e+06 -18388887 <2e-16 ***
## A.compar 4.725e+13 2.591e+06 18239123 <2e-16 ***
## A.total 4.396e+14 4.549e+06 96640870 <2e-16 ***
## T.treatment 2.236e+14 6.171e+06 36241083 <2e-16 ***
## A.intraven -2.657e+14 6.607e+06 -40216643 <2e-16 ***
## T.tamoxifen 5.234e+14 7.660e+06 68329371 <2e-16 ***
## T.result 1.332e+14 7.786e+06 17113118 <2e-16 ***
## A.cours -1.678e+14 4.025e+06 -41691295 <2e-16 ***
## A.observ -4.138e+13 3.127e+06 -13231714 <2e-16 ***
## A.advers -4.201e+14 6.098e+06 -68887349 <2e-16 ***
## A.evalu -1.128e+14 3.098e+06 -36401764 <2e-16 ***
## A.versus 3.139e+13 2.874e+06 10922103 <2e-16 ***
## A.pretreat -1.375e+14 5.479e+06 -25096724 <2e-16 ***
## A.point -1.953e+14 8.331e+06 -23447728 <2e-16 ***
## T.postmenopaus 8.525e+14 1.053e+07 80979490 <2e-16 ***
## A.includ -2.991e+14 4.033e+06 -74170354 <2e-16 ***
## A.seven -3.053e+14 9.164e+06 -33309659 <2e-16 ***
## A.nausea -3.184e+14 6.502e+06 -48977772 <2e-16 ***
## .rnorm 7.743e+12 2.213e+06 3499816 <2e-16 ***
## A.five 2.092e+14 7.103e+06 29453713 <2e-16 ***
## A.better -2.000e+14 6.686e+06 -29917934 <2e-16 ***
## A.multicent -4.818e+14 9.277e+06 -51931305 <2e-16 ***
## A.oper -1.608e+14 6.868e+06 -23407785 <2e-16 ***
## A.stage 6.237e+14 4.864e+06 128230772 <2e-16 ***
## A.alon 6.991e+13 3.297e+06 21203898 <2e-16 ***
## A.less -6.018e+13 4.135e+06 -14553298 <2e-16 ***
## A.random 3.457e+13 3.296e+06 10486363 <2e-16 ***
## A.although -9.994e+13 7.561e+06 -13217813 <2e-16 ***
## A.addit 5.929e+13 4.080e+06 14533260 <2e-16 ***
## A.agent 3.549e+13 5.136e+06 6909383 <2e-16 ***
## A.due 1.335e+14 6.782e+06 19686879 <2e-16 ***
## A.andor 2.307e+14 7.611e+06 30310331 <2e-16 ***
## A.lymph 3.632e+14 5.694e+06 63788692 <2e-16 ***
## A.followup 1.915e+14 4.731e+06 40474440 <2e-16 ***
## A.postmenopaus -5.214e+13 4.213e+06 -12374274 <2e-16 ***
## A.involv 1.454e+14 6.472e+06 22460699 <2e-16 ***
## A.side 2.866e+14 6.897e+06 41547776 <2e-16 ***
## A.standard 3.811e+14 4.217e+06 90370995 <2e-16 ***
## A.receptorposit -5.402e+13 7.133e+06 -7573744 <2e-16 ***
## A.conduct -1.431e+14 7.763e+06 -18437867 <2e-16 ***
## A.neoadjuv 5.357e+13 4.238e+06 12641440 <2e-16 ***
## A.major 4.615e+14 9.214e+06 50086830 <2e-16 ***
## A.administr 1.784e+14 4.903e+06 36384997 <2e-16 ***
## A.limit -1.354e+14 7.535e+06 -17970609 <2e-16 ***
## A.consist 1.844e+14 6.932e+06 26605054 <2e-16 ***
## A.differ -6.919e+13 2.842e+06 -24344196 <2e-16 ***
## A.num.words.unq 1.347e+11 2.791e+05 482535 <2e-16 ***
## A.singl 7.343e+13 7.365e+06 9970896 <2e-16 ***
## A.hazard 1.767e+14 6.933e+06 25485442 <2e-16 ***
## A.sever -5.174e+13 4.771e+06 -10843731 <2e-16 ***
## A.result 1.816e+14 3.386e+06 53631641 <2e-16 ***
## A.randomis -7.001e+13 5.781e+06 -12109300 <2e-16 ***
## A.appear 2.792e+14 6.978e+06 40007149 <2e-16 ***
## A.treat 2.857e+13 2.884e+06 9906343 <2e-16 ***
## A.profil -5.183e+14 6.493e+06 -79821955 <2e-16 ***
## A.year 3.681e+13 2.277e+06 16167236 <2e-16 ***
## A.accord 1.301e+14 6.814e+06 19094475 <2e-16 ***
## T.patient 9.910e+13 5.376e+06 18435118 <2e-16 ***
## A.site -1.838e+14 6.335e+06 -29005985 <2e-16 ***
## A.improv 4.446e+13 3.531e+06 12591024 <2e-16 ***
## A.tamoxifen 4.265e+13 1.664e+06 25627089 <2e-16 ***
## A.ratio -1.106e+14 6.374e+06 -17359038 <2e-16 ***
## A.evid 4.230e+14 7.527e+06 56200924 <2e-16 ***
## A.iii -2.290e+14 6.398e+06 -35789524 <2e-16 ***
## A.conclus 2.324e+14 7.032e+06 33042493 <2e-16 ***
## A.demonstr 1.242e+14 5.564e+06 22323286 <2e-16 ***
## A.statist -3.961e+13 4.438e+06 -8924588 <2e-16 ***
## A.among -6.139e+12 4.142e+06 -1481857 <2e-16 ***
## A.relaps -1.539e+14 3.942e+06 -39043757 <2e-16 ***
## A.human 7.194e+13 9.300e+06 7735563 <2e-16 ***
## A.without -8.741e+13 4.825e+06 -18116549 <2e-16 ***
## A.surgeri 2.120e+14 4.038e+06 52493981 <2e-16 ***
## A.drug 8.070e+13 4.064e+06 19859574 <2e-16 ***
## A.infus -1.325e+14 4.606e+06 -28763890 <2e-16 ***
## T.therapi 1.449e+14 6.986e+06 20741773 <2e-16 ***
## A.higher -8.319e+13 4.347e+06 -19138752 <2e-16 ***
## A.purpos -1.003e+15 6.895e+06 -145469293 <2e-16 ***
## A.premenopaus 2.710e+14 4.169e+06 65004854 <2e-16 ***
## A.num.chars 1.417e+12 3.468e+04 40874544 <2e-16 ***
## A.background -6.032e+14 7.271e+06 -82956858 <2e-16 ***
## A.benefit -2.049e+14 3.918e+06 -52284606 <2e-16 ***
## A.growth 2.182e+13 6.889e+06 3167317 <2e-16 ***
## T.chemotherapi 4.491e+14 6.577e+06 68292598 <2e-16 ***
## A.confirm 2.857e+14 7.156e+06 39926297 <2e-16 ***
## A.inhibitor 4.437e+14 1.019e+07 43543877 <2e-16 ***
## A.characterist -2.665e+14 8.746e+06 -30472157 <2e-16 ***
## A.whether -7.848e+13 6.459e+06 -12150330 <2e-16 ***
## A.popul -1.776e+14 5.856e+06 -30324522 <2e-16 ***
## A.initi 5.306e+13 5.176e+06 10251305 <2e-16 ***
## A.group 2.712e+13 1.175e+06 23084255 <2e-16 ***
## A.mastectomi 1.048e+14 6.579e+06 15923563 <2e-16 ***
## A.least -3.809e+14 6.953e+06 -54781669 <2e-16 ***
## A.consid 4.746e+14 7.876e+06 60259337 <2e-16 ***
## A.node -7.375e+13 4.624e+06 -15949723 <2e-16 ***
## A.axillari -2.569e+14 3.899e+06 -65887362 <2e-16 ***
## A.num.words -2.717e+13 4.389e+05 -61899060 <2e-16 ***
## A.caus -4.010e+14 8.677e+06 -46211327 <2e-16 ***
## A.endocrin -2.064e+14 3.778e+06 -54644055 <2e-16 ***
## A.estim -1.800e+14 7.074e+06 -25451133 <2e-16 ***
## A.new -3.051e+14 5.781e+06 -52783342 <2e-16 ***
## A.postop 1.826e+14 5.091e+06 35870174 <2e-16 ***
## A.distant 2.644e+14 7.403e+06 35715152 <2e-16 ***
## A.regard 1.194e+14 8.793e+06 13581613 <2e-16 ***
## A.requir 4.048e+13 6.860e+06 5899896 <2e-16 ***
## A.defin 4.184e+14 8.849e+06 47284192 <2e-16 ***
## A.trend 1.590e+14 7.868e+06 20207325 <2e-16 ***
## A.subgroup -1.285e+14 5.786e+06 -22208493 <2e-16 ***
## A.receptor 1.018e+14 4.739e+06 21475781 <2e-16 ***
## A.greater 2.317e+14 4.966e+06 46658048 <2e-16 ***
## A.therapi -4.826e+13 2.003e+06 -24091656 <2e-16 ***
## A.studi 7.786e+13 2.136e+06 36455067 <2e-16 ***
## A.howev 5.285e+14 5.691e+06 92860072 <2e-16 ***
## A.support -3.870e+14 5.058e+06 -76516051 <2e-16 ***
## T.adjuv 2.973e+14 7.030e+06 42288459 <2e-16 ***
## A.number -1.389e+14 4.592e+06 -30256656 <2e-16 ***
## A.develop -1.019e+13 5.071e+06 -2009793 <2e-16 ***
## A.comparison -2.882e+14 8.255e+06 -34909085 <2e-16 ***
## A.prospect 1.515e+14 6.402e+06 23660818 <2e-16 ***
## A.regress 2.240e+14 8.470e+06 26451683 <2e-16 ***
## A.per 2.308e+13 4.968e+06 4645388 <2e-16 ***
## A.age 4.708e+13 3.758e+06 12528389 <2e-16 ***
## A.general -3.611e+14 7.547e+06 -47849717 <2e-16 ***
## A.earli -2.164e+14 4.251e+06 -50899826 <2e-16 ***
## A.X001 6.532e+13 7.027e+06 9295494 <2e-16 ***
## A.base -3.551e+14 8.452e+06 -42013664 <2e-16 ***
## A.carcinoma 1.027e+14 4.866e+06 21099241 <2e-16 ***
## A.method 3.748e+14 5.587e+06 67077993 <2e-16 ***
## A.patholog 3.623e+13 4.465e+06 8114162 <2e-16 ***
## A.set 1.703e+14 6.028e+06 28248285 <2e-16 ***
## A.her2 4.177e+13 2.715e+06 15384474 <2e-16 ***
## A.status -2.458e+14 3.606e+06 -68160465 <2e-16 ***
## A.start -3.127e+14 7.486e+06 -41772165 <2e-16 ***
## A.incid -2.730e+14 4.529e+06 -60263221 <2e-16 ***
## A.determin -2.530e+14 5.821e+06 -43472889 <2e-16 ***
## A.posit 1.126e+14 3.691e+06 30509331 <2e-16 ***
## A.progesteron 3.352e+14 9.655e+06 34717175 <2e-16 ***
## A.aromatas -4.674e+14 1.084e+07 -43095389 <2e-16 ***
## A.local -3.326e+14 4.562e+06 -72896492 <2e-16 ***
## A.main -9.813e+13 7.965e+06 -12321000 <2e-16 ***
## A.present -4.521e+14 5.790e+06 -78081904 <2e-16 ***
## A.aim 9.880e+13 7.084e+06 13948392 <2e-16 ***
## A.possibl -1.082e+14 7.445e+06 -14535332 <2e-16 ***
## A.perform -1.127e+14 4.970e+06 -22673512 <2e-16 ***
## A.larg -3.069e+13 8.691e+06 -3531023 <2e-16 ***
## A.shown -4.075e+14 8.443e+06 -48264487 <2e-16 ***
## A.hormon -4.003e+13 3.879e+06 -10319470 <2e-16 ***
## A.period -1.087e+13 7.236e+06 -1501933 <2e-16 ***
## A.reduct 1.805e+14 4.457e+06 40490465 <2e-16 ***
## T.clinic 1.797e+14 9.055e+06 19848043 <2e-16 ***
## A.second 1.850e+14 7.759e+06 23845938 <2e-16 ***
## A.breast -1.996e+14 2.760e+06 -72303857 <2e-16 ***
## A.estrogen -3.447e+14 4.277e+06 -80599383 <2e-16 ***
## A.cancer 3.065e+14 2.505e+06 122393484 <2e-16 ***
## A.tissu 2.981e+14 4.536e+06 65716866 <2e-16 ***
## A.adjuv 4.351e+13 2.620e+06 16603621 <2e-16 ***
## A.case 4.422e+13 4.086e+06 10822515 <2e-16 ***
## A.find 3.164e+14 7.118e+06 44452352 <2e-16 ***
## A.obtain -4.939e+14 7.821e+06 -63156699 <2e-16 ***
## A.suggest 2.728e+14 6.148e+06 44366715 <2e-16 ***
## A.within -1.812e+14 6.667e+06 -27183673 <2e-16 ***
## A.remain -2.793e+14 7.268e+06 -38434784 <2e-16 ***
## A.design 3.653e+13 6.743e+06 5417267 <2e-16 ***
## A.X0001 -1.017e+14 4.266e+06 -23832738 <2e-16 ***
## A.potenti -4.364e+14 7.987e+06 -54632212 <2e-16 ***
## A.import -1.577e+14 8.769e+06 -17985420 <2e-16 ***
## A.detect -2.242e+14 6.525e+06 -34356422 <2e-16 ***
## A.women 6.899e+13 2.165e+06 31863769 <2e-16 ***
## A.recurr -6.834e+13 3.521e+06 -19411713 <2e-16 ***
## A.investig -6.012e+14 5.387e+06 -111599082 <2e-16 ***
## A.chemotherapi -1.699e+14 1.626e+06 -104502777 <2e-16 ***
## A.histolog -7.611e+14 7.808e+06 -97483209 <2e-16 ***
## A.lower 1.225e+14 6.721e+06 18224852 <2e-16 ***
## T.earli -3.988e+14 1.014e+07 -39319914 <2e-16 ***
## A.select 2.354e+14 8.164e+06 28836036 <2e-16 ***
## A.first 6.417e+13 3.889e+06 16501143 <2e-16 ***
## A.signific -8.553e+13 2.258e+06 -37875226 <2e-16 ***
## A.low 2.775e+14 5.970e+06 46493173 <2e-16 ***
## A.mean 1.246e+14 4.411e+06 28237766 <2e-16 ***
## T.women -7.524e+14 8.485e+06 -88676852 <2e-16 ***
## A.radiotherapi 7.979e+13 3.458e+06 23073330 <2e-16 ***
## A.clinic 8.540e+13 3.030e+06 28179919 <2e-16 ***
## A.also -1.063e+14 4.844e+06 -21937064 <2e-16 ***
## A.relat -1.804e+14 4.687e+06 -38483664 <2e-16 ***
## A.life 7.479e+14 9.912e+06 75460723 <2e-16 ***
## A.size -4.946e+13 5.572e+06 -8875681 <2e-16 ***
## A.effect 5.692e+13 2.673e+06 21297450 <2e-16 ***
## A.outcom -2.844e+14 5.022e+06 -56634848 <2e-16 ***
## A.wherea -5.610e+14 6.948e+06 -80738707 <2e-16 ***
## A.may 1.638e+14 4.580e+06 35764919 <2e-16 ***
## A.indic 1.452e+14 5.667e+06 25626195 <2e-16 ***
## A.system 1.608e+14 5.165e+06 31129617 <2e-16 ***
## A.test 8.058e+13 4.980e+06 16181397 <2e-16 ***
## A.metastas 2.228e+13 3.647e+06 6109321 <2e-16 ***
## A.type 2.972e+14 7.368e+06 40340934 <2e-16 ***
## A.function. -1.263e+14 5.378e+06 -23478207 <2e-16 ***
## A.analysi -6.862e+12 4.057e+06 -1691315 <2e-16 ***
## A.avail 5.381e+14 9.366e+06 57444560 <2e-16 ***
## A.negat -6.693e+13 4.993e+06 -13406128 <2e-16 ***
## A.analys -1.571e+14 6.692e+06 -23482248 <2e-16 ***
## A.X005 -3.449e+14 7.169e+06 -48108724 <2e-16 ***
## A.report -3.885e+14 4.590e+06 -84634795 <2e-16 ***
## A.show 9.591e+13 3.850e+06 24911468 <2e-16 ***
## A.provid -4.431e+14 7.781e+06 -56947891 <2e-16 ***
## A.tumor 1.222e+14 2.075e+06 58886312 <2e-16 ***
## A.qualiti -5.827e+14 9.353e+06 -62300965 <2e-16 ***
## A.can -8.007e+13 6.807e+06 -11764046 <2e-16 ***
## A.doubleblind -9.535e+14 9.433e+06 -101080279 <2e-16 ***
## A.tumour -5.982e+13 2.963e+06 -20192469 <2e-16 ***
## A.serum -1.582e+13 2.791e+06 -5667798 <2e-16 ***
## A.factor 6.593e+13 4.122e+06 15993045 <2e-16 ***
## A.analyz -3.472e+13 8.158e+06 -4255887 <2e-16 ***
## A.proport -2.947e+13 8.030e+06 -3669904 <2e-16 ***
## A.valu 1.092e+14 4.973e+06 21963632 <2e-16 ***
## A.sampl -9.678e+11 5.665e+06 -170831 <2e-16 ***
## A.inform 1.535e+14 9.908e+06 15488393 <2e-16 ***
## A.assess 1.471e+14 3.139e+06 46860776 <2e-16 ***
## A.need -4.880e+14 8.210e+06 -59440743 <2e-16 ***
## A.independ 1.913e+14 7.301e+06 26197176 <2e-16 ***
## A.correl -5.802e+13 5.355e+06 -10834768 <2e-16 ***
## A.increas -1.025e+14 2.901e+06 -35322069 <2e-16 ***
## A.level -1.595e+13 2.214e+06 -7205743 <2e-16 ***
## A.placebo 1.431e+14 2.879e+06 49709058 <2e-16 ***
## A.multivari -4.133e+14 9.446e+06 -43751676 <2e-16 ***
## A.model 5.780e+13 6.859e+06 8426933 <2e-16 ***
## A.trial -1.784e+14 2.406e+06 -74163437 <2e-16 ***
## A.high -8.466e+13 4.280e+06 -19781258 <2e-16 ***
## A.score -2.678e+14 3.963e+06 -67576125 <2e-16 ***
## A.express -1.421e+13 3.066e+06 -4634161 <2e-16 ***
## A.found -1.129e+14 5.581e+06 -20225692 <2e-16 ***
## A.particip -2.445e+13 7.148e+06 -3420077 <2e-16 ***
## A.control -1.415e+14 2.713e+06 -52169702 <2e-16 ***
## A.examin -5.062e+14 6.375e+06 -79402119 <2e-16 ***
## A.prognost -1.138e+14 5.771e+06 -19714653 <2e-16 ***
## A.marker -2.626e+14 4.759e+06 -55185037 <2e-16 ***
## A.reduc -2.849e+14 4.319e+06 -65970545 <2e-16 ***
## A.cell -1.792e+14 3.944e+06 -45435133 <2e-16 ***
## A.measur -3.546e+13 3.734e+06 -9496232 <2e-16 ***
## A.chang -1.100e+14 3.696e+06 -29759237 <2e-16 ***
## T.respons -5.746e+14 1.120e+07 -51304342 <2e-16 ***
## A.baselin 2.958e+13 4.482e+06 6600306 <2e-16 ***
## A.identifi -2.394e+14 7.256e+06 -32993697 <2e-16 ***
## A.decreas 2.005e+14 4.236e+06 47328513 <2e-16 ***
## A.bone -1.100e+14 2.512e+06 -43787720 <2e-16 ***
## A.prevent -2.846e+14 6.427e+06 -44275918 <2e-16 ***
## A.data -1.169e+14 4.303e+06 -27158227 <2e-16 ***
## A.associ 1.537e+14 3.492e+06 44026628 <2e-16 ***
## A.predict -1.635e+14 4.104e+06 -39848595 <2e-16 ***
## T.effect -6.010e+14 7.722e+06 -77823310 <2e-16 ***
## A.risk -7.290e+13 2.813e+06 -25911561 <2e-16 ***
## A.use -1.337e+14 2.613e+06 -51169643 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1785.7 on 1301 degrees of freedom
## Residual deviance: 5478.6 on 928 degrees of freedom
## AIC: 6226.6
##
## Number of Fisher Scoring iterations: 25
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.glm.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Conditional.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.9342561
## 3 0.2 0.9342561
## 4 0.3 0.9342561
## 5 0.4 0.9342561
## 6 0.5 0.9342561
## 7 0.6 0.9342561
## 8 0.7 0.9342561
## 9 0.8 0.9342561
## 10 0.9 0.9342561
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.9000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Reference
## Prediction N Y
## N 686 32
## Y 44 540
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 686
## 2 Y 32
## trial.fctr.predict.Conditional.X.glm.Y
## 1 44
## 2 540
## Prediction
## Reference N Y
## N 686 44
## Y 32 540
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.416283e-01 8.817800e-01 9.274808e-01 9.537375e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.877499e-211 2.070256e-01
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.glm.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Conditional.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6910569
## 3 0.2 0.6910569
## 4 0.3 0.6910569
## 5 0.4 0.6910569
## 6 0.5 0.6910569
## 7 0.6 0.6910569
## 8 0.7 0.6910569
## 9 0.8 0.6910569
## 10 0.9 0.6910569
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.9000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Reference
## Prediction N Y
## N 236 75
## Y 77 170
## trial.fctr trial.fctr.predict.Conditional.X.glm.N
## 1 N 236
## 2 Y 75
## trial.fctr.predict.Conditional.X.glm.Y
## 1 77
## 2 170
## Prediction
## Reference N Y
## N 236 77
## Y 75 170
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.275986e-01 4.474744e-01 6.886224e-01 7.641450e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 3.160021e-16 9.353539e-01
## model_id model_method
## 1 Conditional.X.glm glm
## feats
## 1 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 19.673 6.063
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.941891 0.9 0.9342561 0.7058372
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9274808 0.9537375 0.4074552 0.7239356
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.9 0.6910569 0.7275986
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.6886224 0.764145 0.4474744 6226.635
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.0013303 0.002711058
## [1] "fitting model: Conditional.X.rpart"
## [1] " indep_vars: T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use"
## + Fold1: cp=0.02972
## - Fold1: cp=0.02972
## + Fold2: cp=0.02972
## - Fold2: cp=0.02972
## + Fold3: cp=0.02972
## - Fold3: cp=0.02972
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.0297 on full training set## Warning in myfit_mdl(model_id = paste0(model_id_pfx, ""), model_method =
## method, : model's bestTune found at an extreme of tuneGrid for parameter:
## cp
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1302
##
## CP nsplit rel error
## 1 0.35489510 0 1.0000000
## 2 0.10139860 1 0.6451049
## 3 0.02972028 2 0.5437063
##
## Variable importance
## T.phase A.toxic T.iii A.neutropenia
## 42 18 13 8
## A.mbc A.phase A.progressionfre A.hematolog
## 6 6 5 1
## A.mgm2 A.day
## 1 1
##
## Node number 1: 1302 observations, complexity param=0.3548951
## predicted class=N expected loss=0.4393241 P(node) =1
## class counts: 730 572
## probabilities: 0.561 0.439
## left son=2 (1005 obs) right son=3 (297 obs)
## Primary splits:
## T.phase < 0.5 to the left, improve=124.62490, (0 missing)
## A.toxic < 0.5 to the left, improve=117.23600, (0 missing)
## A.mgm2 < 0.5 to the left, improve=107.91430, (0 missing)
## A.respons < 0.5 to the left, improve= 83.18161, (0 missing)
## A.everi < 0.5 to the left, improve= 72.81706, (0 missing)
## Surrogate splits:
## T.iii < 0.5 to the left, agree=0.840, adj=0.300, (0 split)
## A.neutropenia < 0.5 to the left, agree=0.807, adj=0.155, (0 split)
## A.mbc < 0.5 to the left, agree=0.803, adj=0.135, (0 split)
## A.phase < 0.5 to the left, agree=0.802, adj=0.131, (0 split)
## A.progressionfre < 0.5 to the left, agree=0.797, adj=0.111, (0 split)
##
## Node number 2: 1005 observations, complexity param=0.1013986
## predicted class=N expected loss=0.320398 P(node) =0.7718894
## class counts: 683 322
## probabilities: 0.680 0.320
## left son=4 (811 obs) right son=5 (194 obs)
## Primary splits:
## A.toxic < 0.5 to the left, improve=52.07115, (0 missing)
## A.mgm2 < 0.5 to the left, improve=49.36527, (0 missing)
## A.surviv < 0.5 to the left, improve=48.36917, (0 missing)
## T.versus < 0.5 to the left, improve=45.55312, (0 missing)
## A.cyclophosphamid < 0.5 to the left, improve=41.05959, (0 missing)
## Surrogate splits:
## A.neutropenia < 0.5 to the left, agree=0.824, adj=0.088, (0 split)
## A.hematolog < 0.5 to the left, agree=0.821, adj=0.072, (0 split)
## A.mgm2 < 3.5 to the left, agree=0.819, adj=0.062, (0 split)
## A.day < 4.5 to the left, agree=0.814, adj=0.036, (0 split)
## T.docetaxel < 1.5 to the left, agree=0.811, adj=0.021, (0 split)
##
## Node number 3: 297 observations
## predicted class=Y expected loss=0.1582492 P(node) =0.2281106
## class counts: 47 250
## probabilities: 0.158 0.842
##
## Node number 4: 811 observations
## predicted class=N expected loss=0.2416769 P(node) =0.6228879
## class counts: 615 196
## probabilities: 0.758 0.242
##
## Node number 5: 194 observations
## predicted class=Y expected loss=0.3505155 P(node) =0.1490015
## class counts: 68 126
## probabilities: 0.351 0.649
##
## n= 1302
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1302 572 N (0.5606759 0.4393241)
## 2) T.phase< 0.5 1005 322 N (0.6796020 0.3203980)
## 4) A.toxic< 0.5 811 196 N (0.7583231 0.2416769) *
## 5) A.toxic>=0.5 194 68 Y (0.3505155 0.6494845) *
## 3) T.phase>=0.5 297 47 Y (0.1582492 0.8417508) *
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 615 196
## Y 115 376
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 615
## 2 Y 196
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 115
## 2 376
## Reference
## Prediction N Y
## N 615 196
## Y 115 376
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 615
## 2 Y 196
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 115
## 2 376
## Reference
## Prediction N Y
## N 615 196
## Y 115 376
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 615
## 2 Y 196
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 115
## 2 376
## Reference
## Prediction N Y
## N 615 196
## Y 115 376
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 615
## 2 Y 196
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 115
## 2 376
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 683 322
## Y 47 250
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 683
## 2 Y 322
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 47
## 2 250
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6104589
## 2 0.1 0.6104589
## 3 0.2 0.6104589
## 4 0.3 0.7074318
## 5 0.4 0.7074318
## 6 0.5 0.7074318
## 7 0.6 0.7074318
## 8 0.7 0.5753740
## 9 0.8 0.5753740
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.6000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 615
## 2 Y 196
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 115
## 2 376
## Reference
## Prediction N Y
## N 615 196
## Y 115 376
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 615
## 2 Y 196
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 115
## 2 376
## Prediction
## Reference N Y
## N 615 115
## Y 196 376
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.611367e-01 5.075871e-01 7.370177e-01 7.840707e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 2.401266e-51 5.722699e-06
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 270 88
## Y 43 157
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 270
## 2 Y 88
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 43
## 2 157
## Reference
## Prediction N Y
## N 270 88
## Y 43 157
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 270
## 2 Y 88
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 43
## 2 157
## Reference
## Prediction N Y
## N 270 88
## Y 43 157
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 270
## 2 Y 88
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 43
## 2 157
## Reference
## Prediction N Y
## N 270 88
## Y 43 157
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 270
## 2 Y 88
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 43
## 2 157
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 298 148
## Y 15 97
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 298
## 2 Y 148
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 15
## 2 97
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6102117
## 3 0.2 0.6102117
## 4 0.3 0.7056180
## 5 0.4 0.7056180
## 6 0.5 0.7056180
## 7 0.6 0.7056180
## 8 0.7 0.5434174
## 9 0.8 0.5434174
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.6000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 270
## 2 Y 88
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 43
## 2 157
## Reference
## Prediction N Y
## N 270 88
## Y 43 157
## trial.fctr trial.fctr.predict.Conditional.X.rpart.N
## 1 N 270
## 2 Y 88
## trial.fctr.predict.Conditional.X.rpart.Y
## 1 43
## 2 157
## Prediction
## Reference N Y
## N 270 43
## Y 88 157
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.652330e-01 5.136851e-01 7.278160e-01 7.998142e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 6.865825e-24 1.208981e-04
## model_id model_method
## 1 Conditional.X.rpart rpart
## feats
## 1 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 7.181 1.182
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.7631694 0.6 0.7074318 0.7588326
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7370177 0.7840707 0.5017505 0.7635587
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.6 0.705618 0.765233
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.727816 0.7998142 0.5136851
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.01350107 0.02864154
## [1] "fitting model: Conditional.X.cp.0.rpart"
## [1] " indep_vars: T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use"
## Fitting cp = 0 on full training set
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1302
##
## CP nsplit rel error
## 1 0.3548951049 0 1.0000000
## 2 0.1013986014 1 0.6451049
## 3 0.0297202797 2 0.5437063
## 4 0.0139860140 3 0.5139860
## 5 0.0134032634 4 0.5000000
## 6 0.0104895105 9 0.4125874
## 7 0.0087412587 10 0.4020979
## 8 0.0052447552 15 0.3583916
## 9 0.0029137529 19 0.3374126
## 10 0.0026223776 22 0.3286713
## 11 0.0017482517 26 0.3181818
## 12 0.0006993007 36 0.2972028
## 13 0.0000000000 41 0.2937063
##
## Variable importance
## T.phase A.toxic T.iii A.surviv
## 18 8 5 4
## A.neutropenia A.mbc A.phase A.mgm2
## 3 2 2 2
## A.predict A.progressionfre A.reduc A.overal
## 2 2 2 2
## A.cycl A.num.words A.num.words.unq A.median
## 2 2 1 1
## A.diseasefre T.versus T.breast A.efficaci
## 1 1 1 1
## A.durat A.firstlin A.followup A.chang
## 1 1 1 1
## A.num.chars T.num.chars A.control A.patient
## 1 1 1 1
## A.postop A.data T.num.words.unq T.num.words
## 1 1 1 1
## A.month A.benefit A.improv A.diseas
## 1 1 1 1
## A.without A.chemotherapi A.treat A.arm
## 1 1 1 1
## A.hematolog T.studi A.life
## 1 1 1
##
## Node number 1: 1302 observations, complexity param=0.3548951
## predicted class=N expected loss=0.4393241 P(node) =1
## class counts: 730 572
## probabilities: 0.561 0.439
## left son=2 (1005 obs) right son=3 (297 obs)
## Primary splits:
## T.phase < 0.5 to the left, improve=124.62490, (0 missing)
## A.toxic < 0.5 to the left, improve=117.23600, (0 missing)
## A.mgm2 < 0.5 to the left, improve=107.91430, (0 missing)
## A.respons < 0.5 to the left, improve= 83.18161, (0 missing)
## A.everi < 0.5 to the left, improve= 72.81706, (0 missing)
## Surrogate splits:
## T.iii < 0.5 to the left, agree=0.840, adj=0.300, (0 split)
## A.neutropenia < 0.5 to the left, agree=0.807, adj=0.155, (0 split)
## A.mbc < 0.5 to the left, agree=0.803, adj=0.135, (0 split)
## A.phase < 0.5 to the left, agree=0.802, adj=0.131, (0 split)
## A.progressionfre < 0.5 to the left, agree=0.797, adj=0.111, (0 split)
##
## Node number 2: 1005 observations, complexity param=0.1013986
## predicted class=N expected loss=0.320398 P(node) =0.7718894
## class counts: 683 322
## probabilities: 0.680 0.320
## left son=4 (811 obs) right son=5 (194 obs)
## Primary splits:
## A.toxic < 0.5 to the left, improve=52.07115, (0 missing)
## A.mgm2 < 0.5 to the left, improve=49.36527, (0 missing)
## A.surviv < 0.5 to the left, improve=48.36917, (0 missing)
## T.versus < 0.5 to the left, improve=45.55312, (0 missing)
## A.cyclophosphamid < 0.5 to the left, improve=41.05959, (0 missing)
## Surrogate splits:
## A.neutropenia < 0.5 to the left, agree=0.824, adj=0.088, (0 split)
## A.hematolog < 0.5 to the left, agree=0.821, adj=0.072, (0 split)
## A.mgm2 < 3.5 to the left, agree=0.819, adj=0.062, (0 split)
## A.day < 4.5 to the left, agree=0.814, adj=0.036, (0 split)
## T.docetaxel < 1.5 to the left, agree=0.811, adj=0.021, (0 split)
##
## Node number 3: 297 observations, complexity param=0.01048951
## predicted class=Y expected loss=0.1582492 P(node) =0.2281106
## class counts: 47 250
## probabilities: 0.158 0.842
## left son=6 (16 obs) right son=7 (281 obs)
## Primary splits:
## T.breast < 0.5 to the left, improve=9.473778, (0 missing)
## A.risk < 1.5 to the right, improve=7.003397, (0 missing)
## A.improv < 2.5 to the right, improve=5.757797, (0 missing)
## A.respons < 0.5 to the left, improve=5.190839, (0 missing)
## A.life < 0.5 to the right, improve=5.016553, (0 missing)
## Surrogate splits:
## A.infus < 3.5 to the right, agree=0.953, adj=0.125, (0 split)
## A.administ < 2.5 to the right, agree=0.949, adj=0.062, (0 split)
## A.use < 2.5 to the right, agree=0.949, adj=0.062, (0 split)
##
## Node number 4: 811 observations, complexity param=0.01340326
## predicted class=N expected loss=0.2416769 P(node) =0.6228879
## class counts: 615 196
## probabilities: 0.758 0.242
## left son=8 (484 obs) right son=9 (327 obs)
## Primary splits:
## A.surviv < 0.5 to the left, improve=26.62655, (0 missing)
## T.versus < 0.5 to the left, improve=25.71859, (0 missing)
## A.median < 0.5 to the left, improve=25.35863, (0 missing)
## A.cyclophosphamid < 0.5 to the left, improve=20.63959, (0 missing)
## A.mgm2 < 0.5 to the left, improve=17.22566, (0 missing)
## Surrogate splits:
## A.overal < 0.5 to the left, agree=0.771, adj=0.431, (0 split)
## A.diseasefre < 0.5 to the left, agree=0.732, adj=0.336, (0 split)
## A.median < 0.5 to the left, agree=0.695, adj=0.245, (0 split)
## A.followup < 0.5 to the left, agree=0.688, adj=0.226, (0 split)
## A.benefit < 0.5 to the left, agree=0.672, adj=0.187, (0 split)
##
## Node number 5: 194 observations, complexity param=0.02972028
## predicted class=Y expected loss=0.3505155 P(node) =0.1490015
## class counts: 68 126
## probabilities: 0.351 0.649
## left son=10 (33 obs) right son=11 (161 obs)
## Primary splits:
## A.reduc < 0.5 to the right, improve=13.177630, (0 missing)
## A.chang < 0.5 to the right, improve=11.126450, (0 missing)
## A.respons < 0.5 to the left, improve=10.170190, (0 missing)
## A.control < 0.5 to the right, improve= 9.552341, (0 missing)
## A.mgm2 < 2.5 to the left, improve= 9.441306, (0 missing)
## Surrogate splits:
## A.placebo < 0.5 to the right, agree=0.851, adj=0.121, (0 split)
## A.cell < 3.5 to the right, agree=0.840, adj=0.061, (0 split)
## A.week < 6.5 to the right, agree=0.835, adj=0.030, (0 split)
## A.day < 6.5 to the right, agree=0.835, adj=0.030, (0 split)
## A.dose < 8.5 to the right, agree=0.835, adj=0.030, (0 split)
##
## Node number 6: 16 observations
## predicted class=N expected loss=0.3125 P(node) =0.01228879
## class counts: 11 5
## probabilities: 0.688 0.312
##
## Node number 7: 281 observations, complexity param=0.005244755
## predicted class=Y expected loss=0.1281139 P(node) =0.2158218
## class counts: 36 245
## probabilities: 0.128 0.872
## left son=14 (7 obs) right son=15 (274 obs)
## Primary splits:
## A.improv < 2.5 to the right, improve=4.933256, (0 missing)
## A.life < 0.5 to the right, improve=4.590382, (0 missing)
## A.mgm2 < 0.5 to the left, improve=4.398620, (0 missing)
## A.baselin < 0.5 to the right, improve=4.369834, (0 missing)
## A.qualiti < 0.5 to the right, improve=3.987922, (0 missing)
## Surrogate splits:
## A.X0001 < 2.5 to the right, agree=0.982, adj=0.286, (0 split)
## A.addit < 2.5 to the right, agree=0.979, adj=0.143, (0 split)
##
## Node number 8: 484 observations, complexity param=0.008741259
## predicted class=N expected loss=0.1363636 P(node) =0.3717358
## class counts: 418 66
## probabilities: 0.864 0.136
## left son=16 (475 obs) right son=17 (9 obs)
## Primary splits:
## A.firstlin < 0.5 to the left, improve=7.545731, (0 missing)
## T.versus < 0.5 to the left, improve=7.172498, (0 missing)
## A.mgm2 < 0.5 to the left, improve=5.810548, (0 missing)
## A.respons < 0.5 to the left, improve=4.732526, (0 missing)
## A.median < 0.5 to the left, improve=4.694444, (0 missing)
## Surrogate splits:
## A.mbc < 1.5 to the left, agree=0.983, adj=0.111, (0 split)
##
## Node number 9: 327 observations, complexity param=0.01340326
## predicted class=N expected loss=0.3975535 P(node) =0.2511521
## class counts: 197 130
## probabilities: 0.602 0.398
## left son=18 (55 obs) right son=19 (272 obs)
## Primary splits:
## A.predict < 0.5 to the right, improve=15.558950, (0 missing)
## T.versus < 0.5 to the left, improve=12.052600, (0 missing)
## T.respons < 0.5 to the right, improve= 9.900773, (0 missing)
## A.toler < 0.5 to the left, improve= 9.480157, (0 missing)
## A.associ < 0.5 to the right, improve= 9.328749, (0 missing)
## Surrogate splits:
## A.prognost < 1.5 to the right, agree=0.856, adj=0.145, (0 split)
## A.multivari < 0.5 to the right, agree=0.853, adj=0.127, (0 split)
## A.marker < 1.5 to the right, agree=0.850, adj=0.109, (0 split)
## A.identifi < 0.5 to the right, agree=0.850, adj=0.109, (0 split)
## A.patient < 11.5 to the right, agree=0.847, adj=0.091, (0 split)
##
## Node number 10: 33 observations, complexity param=0.002622378
## predicted class=N expected loss=0.2424242 P(node) =0.02534562
## class counts: 25 8
## probabilities: 0.758 0.242
## left son=20 (12 obs) right son=21 (21 obs)
## Primary splits:
## A.clinic < 0.5 to the right, improve=2.216450, (0 missing)
## A.treatment < 2.5 to the left, improve=1.923410, (0 missing)
## A.respons < 1.5 to the left, improve=1.903821, (0 missing)
## A.day < 0.5 to the left, improve=1.685122, (0 missing)
## A.howev < 0.5 to the left, improve=1.484848, (0 missing)
## Surrogate splits:
## T.studi < 0.5 to the right, agree=0.788, adj=0.417, (0 split)
## A.can < 0.5 to the right, agree=0.758, adj=0.333, (0 split)
## A.arm < 0.5 to the right, agree=0.727, adj=0.250, (0 split)
## A.patient < 8.5 to the right, agree=0.727, adj=0.250, (0 split)
## A.versus < 0.5 to the right, agree=0.727, adj=0.250, (0 split)
##
## Node number 11: 161 observations, complexity param=0.01398601
## predicted class=Y expected loss=0.2670807 P(node) =0.1236559
## class counts: 43 118
## probabilities: 0.267 0.733
## left son=22 (14 obs) right son=23 (147 obs)
## Primary splits:
## A.chang < 0.5 to the right, improve=8.248743, (0 missing)
## A.control < 0.5 to the right, improve=7.712008, (0 missing)
## A.respons < 0.5 to the left, improve=6.314374, (0 missing)
## A.trial < 1.5 to the right, improve=6.262940, (0 missing)
## A.analyz < 0.5 to the right, improve=6.047834, (0 missing)
## Surrogate splits:
## A.baselin < 0.5 to the right, agree=0.944, adj=0.357, (0 split)
## A.decreas < 2.5 to the right, agree=0.925, adj=0.143, (0 split)
## A.data < 2.5 to the right, agree=0.925, adj=0.143, (0 split)
## T.doxorubicin < 1.5 to the right, agree=0.919, adj=0.071, (0 split)
## A.vomit < 1.5 to the right, agree=0.919, adj=0.071, (0 split)
##
## Node number 14: 7 observations
## predicted class=N expected loss=0.2857143 P(node) =0.005376344
## class counts: 5 2
## probabilities: 0.714 0.286
##
## Node number 15: 274 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.1131387 P(node) =0.2104455
## class counts: 31 243
## probabilities: 0.113 0.887
## left son=30 (12 obs) right son=31 (262 obs)
## Primary splits:
## A.life < 0.5 to the right, improve=3.756394, (0 missing)
## A.mgm2 < 0.5 to the left, improve=3.711647, (0 missing)
## A.trial < 2.5 to the right, improve=3.642844, (0 missing)
## A.placebo < 1.5 to the right, improve=3.313135, (0 missing)
## A.qualiti < 0.5 to the right, improve=3.106614, (0 missing)
## Surrogate splits:
## A.qualiti < 0.5 to the right, agree=0.985, adj=0.667, (0 split)
## A.function. < 1.5 to the right, agree=0.960, adj=0.083, (0 split)
## A.trial < 3.5 to the right, agree=0.960, adj=0.083, (0 split)
##
## Node number 16: 475 observations, complexity param=0.005244755
## predicted class=N expected loss=0.1242105 P(node) =0.3648233
## class counts: 416 59
## probabilities: 0.876 0.124
## left son=32 (450 obs) right son=33 (25 obs)
## Primary splits:
## A.mgm2 < 0.5 to the left, improve=5.263158, (0 missing)
## A.cyclophosphamid < 0.5 to the left, improve=3.935207, (0 missing)
## T.versus < 0.5 to the left, improve=3.487895, (0 missing)
## A.num.words.unq < 75.5 to the right, improve=3.113132, (0 missing)
## A.num.words < 87 to the right, improve=2.745036, (0 missing)
## Surrogate splits:
## A.doxorubicin < 4.5 to the left, agree=0.962, adj=0.28, (0 split)
## A.arm < 8.5 to the left, agree=0.952, adj=0.08, (0 split)
## A.schedul < 1.5 to the left, agree=0.952, adj=0.08, (0 split)
## T.doxorubicin < 1.5 to the left, agree=0.952, adj=0.08, (0 split)
## A.cours < 2.5 to the left, agree=0.952, adj=0.08, (0 split)
##
## Node number 17: 9 observations
## predicted class=Y expected loss=0.2222222 P(node) =0.006912442
## class counts: 2 7
## probabilities: 0.222 0.778
##
## Node number 18: 55 observations
## predicted class=N expected loss=0.05454545 P(node) =0.0422427
## class counts: 52 3
## probabilities: 0.945 0.055
##
## Node number 19: 272 observations, complexity param=0.01340326
## predicted class=N expected loss=0.4669118 P(node) =0.2089094
## class counts: 145 127
## probabilities: 0.533 0.467
## left son=38 (229 obs) right son=39 (43 obs)
## Primary splits:
## A.cycl < 0.5 to the left, improve=9.225880, (0 missing)
## T.versus < 0.5 to the left, improve=9.191636, (0 missing)
## A.mgm2 < 1.5 to the left, improve=8.883475, (0 missing)
## A.cyclophosphamid < 0.5 to the left, improve=8.506190, (0 missing)
## A.either < 0.5 to the left, improve=8.506190, (0 missing)
## Surrogate splits:
## A.mgm2 < 1.5 to the left, agree=0.860, adj=0.116, (0 split)
## T.docetaxel < 0.5 to the left, agree=0.860, adj=0.116, (0 split)
## A.four < 2.5 to the left, agree=0.860, adj=0.116, (0 split)
## A.postop < 2.5 to the left, agree=0.857, adj=0.093, (0 split)
## A.chemotherapi < 8 to the left, agree=0.857, adj=0.093, (0 split)
##
## Node number 20: 12 observations
## predicted class=N expected loss=0 P(node) =0.00921659
## class counts: 12 0
## probabilities: 1.000 0.000
##
## Node number 21: 21 observations, complexity param=0.002622378
## predicted class=N expected loss=0.3809524 P(node) =0.01612903
## class counts: 13 8
## probabilities: 0.619 0.381
## left son=42 (10 obs) right son=43 (11 obs)
## Primary splits:
## A.treat < 0.5 to the right, improve=3.013853, (0 missing)
## A.num.chars < 1497 to the left, improve=3.013853, (0 missing)
## A.num.words.unq < 97 to the left, improve=2.571429, (0 missing)
## A.num.words < 149.5 to the left, improve=2.571429, (0 missing)
## A.respons < 1 to the left, improve=2.333333, (0 missing)
## Surrogate splits:
## A.num.words.unq < 95 to the left, agree=0.857, adj=0.7, (0 split)
## A.num.chars < 1497 to the left, agree=0.810, adj=0.6, (0 split)
## A.num.words < 116 to the left, agree=0.810, adj=0.6, (0 split)
## A.signific < 0.5 to the left, agree=0.810, adj=0.6, (0 split)
## A.cyclophosphamid < 0.5 to the left, agree=0.762, adj=0.5, (0 split)
##
## Node number 22: 14 observations
## predicted class=N expected loss=0.2142857 P(node) =0.01075269
## class counts: 11 3
## probabilities: 0.786 0.214
##
## Node number 23: 147 observations, complexity param=0.008741259
## predicted class=Y expected loss=0.2176871 P(node) =0.1129032
## class counts: 32 115
## probabilities: 0.218 0.782
## left son=46 (19 obs) right son=47 (128 obs)
## Primary splits:
## A.control < 0.5 to the right, improve=7.475922, (0 missing)
## A.mgm2 < 0.5 to the left, improve=5.072706, (0 missing)
## A.respons < 0.5 to the left, improve=4.098330, (0 missing)
## A.effect < 1.5 to the right, improve=3.331952, (0 missing)
## T.breast < 0.5 to the left, improve=3.328633, (0 missing)
## Surrogate splits:
## A.fluorouracil < 1.5 to the right, agree=0.891, adj=0.158, (0 split)
## A.radiotherapi < 2.5 to the right, agree=0.891, adj=0.158, (0 split)
## A.factor < 2.5 to the right, agree=0.891, adj=0.158, (0 split)
## A.involv < 1.5 to the right, agree=0.884, adj=0.105, (0 split)
## A.treat < 3.5 to the right, agree=0.884, adj=0.105, (0 split)
##
## Node number 30: 12 observations
## predicted class=N expected loss=0.5 P(node) =0.00921659
## class counts: 6 6
## probabilities: 0.500 0.500
##
## Node number 31: 262 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.09541985 P(node) =0.2012289
## class counts: 25 237
## probabilities: 0.095 0.905
## left son=62 (129 obs) right son=63 (133 obs)
## Primary splits:
## A.mgm2 < 0.5 to the left, improve=2.306818, (0 missing)
## A.everi < 0.5 to the left, improve=2.103772, (0 missing)
## A.month < 0.5 to the left, improve=1.945861, (0 missing)
## A.import < 0.5 to the right, improve=1.929008, (0 missing)
## A.chemotherapi < 4.5 to the right, improve=1.929008, (0 missing)
## Surrogate splits:
## A.everi < 0.5 to the left, agree=0.691, adj=0.372, (0 split)
## A.day < 0.5 to the left, agree=0.664, adj=0.318, (0 split)
## A.toxic < 0.5 to the left, agree=0.653, adj=0.295, (0 split)
## A.cycl < 0.5 to the left, agree=0.649, adj=0.287, (0 split)
## A.grade < 0.5 to the left, agree=0.634, adj=0.256, (0 split)
##
## Node number 32: 450 observations, complexity param=0.0006993007
## predicted class=N expected loss=0.1066667 P(node) =0.3456221
## class counts: 402 48
## probabilities: 0.893 0.107
## left son=64 (431 obs) right son=65 (19 obs)
## Primary splits:
## T.versus < 0.5 to the left, improve=3.921436, (0 missing)
## A.X5fluorouracil < 0.5 to the left, improve=3.054817, (0 missing)
## A.num.words < 87 to the right, improve=2.992591, (0 missing)
## A.num.chars < 993 to the right, improve=2.921035, (0 missing)
## A.num.words.unq < 73.5 to the right, improve=2.682753, (0 missing)
##
## Node number 33: 25 observations, complexity param=0.005244755
## predicted class=N expected loss=0.44 P(node) =0.01920123
## class counts: 14 11
## probabilities: 0.560 0.440
## left son=66 (13 obs) right son=67 (12 obs)
## Primary splits:
## T.num.chars < 127.5 to the right, improve=4.435385, (0 missing)
## A.num.words < 153 to the right, improve=2.334706, (0 missing)
## A.respons < 0.5 to the left, improve=2.253333, (0 missing)
## A.num.words.unq < 78.5 to the right, improve=2.253333, (0 missing)
## A.result < 0.5 to the right, improve=2.253333, (0 missing)
## Surrogate splits:
## T.num.words < 9.5 to the right, agree=0.88, adj=0.750, (0 split)
## T.num.words.unq < 9.5 to the right, agree=0.88, adj=0.750, (0 split)
## A.num.words.unq < 72.5 to the right, agree=0.80, adj=0.583, (0 split)
## A.group < 0.5 to the right, agree=0.80, adj=0.583, (0 split)
## A.num.words < 148.5 to the right, agree=0.80, adj=0.583, (0 split)
##
## Node number 38: 229 observations, complexity param=0.01340326
## predicted class=N expected loss=0.4104803 P(node) =0.1758833
## class counts: 135 94
## probabilities: 0.590 0.410
## left son=76 (192 obs) right son=77 (37 obs)
## Primary splits:
## A.durat < 0.5 to the left, improve=8.995516, (0 missing)
## A.toler < 0.5 to the left, improve=7.719920, (0 missing)
## A.efficaci < 0.5 to the left, improve=7.466941, (0 missing)
## T.versus < 0.5 to the left, improve=7.354570, (0 missing)
## A.associ < 0.5 to the right, improve=6.795951, (0 missing)
## Surrogate splits:
## A.therapi < 6.5 to the left, agree=0.856, adj=0.108, (0 split)
## A.toler < 1.5 to the left, agree=0.852, adj=0.081, (0 split)
## A.num.words.unq < 140 to the left, agree=0.852, adj=0.081, (0 split)
## A.remain < 1.5 to the left, agree=0.852, adj=0.081, (0 split)
## T.combin < 1.5 to the left, agree=0.847, adj=0.054, (0 split)
##
## Node number 39: 43 observations, complexity param=0.002622378
## predicted class=Y expected loss=0.2325581 P(node) =0.03302611
## class counts: 10 33
## probabilities: 0.233 0.767
## left son=78 (26 obs) right son=79 (17 obs)
## Primary splits:
## A.arm < 0.5 to the left, improve=3.041145, (0 missing)
## A.median < 0.5 to the left, improve=1.953965, (0 missing)
## A.followup < 0.5 to the left, improve=1.805851, (0 missing)
## A.compar < 1.5 to the left, improve=1.697253, (0 missing)
## A.month < 0.5 to the left, improve=1.473837, (0 missing)
## Surrogate splits:
## T.versus < 0.5 to the left, agree=0.744, adj=0.353, (0 split)
## A.mgm2 < 2.5 to the left, agree=0.721, adj=0.294, (0 split)
## A.follow < 2.5 to the left, agree=0.721, adj=0.294, (0 split)
## A.metastat < 0.5 to the left, agree=0.698, adj=0.235, (0 split)
## A.four < 2 to the left, agree=0.698, adj=0.235, (0 split)
##
## Node number 42: 10 observations
## predicted class=N expected loss=0.1 P(node) =0.007680492
## class counts: 9 1
## probabilities: 0.900 0.100
##
## Node number 43: 11 observations
## predicted class=Y expected loss=0.3636364 P(node) =0.008448541
## class counts: 4 7
## probabilities: 0.364 0.636
##
## Node number 46: 19 observations
## predicted class=N expected loss=0.3684211 P(node) =0.01459293
## class counts: 12 7
## probabilities: 0.632 0.368
##
## Node number 47: 128 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.15625 P(node) =0.09831029
## class counts: 20 108
## probabilities: 0.156 0.844
## left son=94 (81 obs) right son=95 (47 obs)
## Primary splits:
## A.mgm2 < 1.5 to the left, improve=2.706133, (0 missing)
## A.analyz < 0.5 to the right, improve=2.552834, (0 missing)
## A.effect < 1.5 to the right, improve=2.353318, (0 missing)
## A.trial < 1.5 to the right, improve=2.331454, (0 missing)
## A.common < 0.5 to the right, improve=2.141608, (0 missing)
## Surrogate splits:
## A.X500 < 0.5 to the left, agree=0.750, adj=0.319, (0 split)
## A.everi < 0.5 to the left, agree=0.719, adj=0.234, (0 split)
## T.result < 0.5 to the left, agree=0.719, adj=0.234, (0 split)
## A.epirubicin < 0.5 to the left, agree=0.703, adj=0.191, (0 split)
## A.neutropenia < 0.5 to the left, agree=0.688, adj=0.149, (0 split)
##
## Node number 62: 129 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.1627907 P(node) =0.09907834
## class counts: 21 108
## probabilities: 0.163 0.837
## left son=124 (68 obs) right son=125 (61 obs)
## Primary splits:
## A.month < 0.5 to the left, improve=2.987284, (0 missing)
## A.carcinoma < 0.5 to the right, improve=2.471924, (0 missing)
## A.chemotherapi < 4.5 to the right, improve=2.471924, (0 missing)
## A.risk < 0.5 to the right, improve=2.471924, (0 missing)
## A.alon < 0.5 to the right, improve=2.427497, (0 missing)
## Surrogate splits:
## A.median < 0.5 to the left, agree=0.775, adj=0.525, (0 split)
## A.metastat < 0.5 to the left, agree=0.713, adj=0.393, (0 split)
## A.surviv < 0.5 to the left, agree=0.698, adj=0.361, (0 split)
## A.progress < 0.5 to the left, agree=0.690, adj=0.344, (0 split)
## A.efficaci < 0.5 to the left, agree=0.690, adj=0.344, (0 split)
##
## Node number 63: 133 observations
## predicted class=Y expected loss=0.03007519 P(node) =0.1021505
## class counts: 4 129
## probabilities: 0.030 0.970
##
## Node number 64: 431 observations, complexity param=0.0006993007
## predicted class=N expected loss=0.09280742 P(node) =0.3310292
## class counts: 391 40
## probabilities: 0.907 0.093
## left son=128 (417 obs) right son=129 (14 obs)
## Primary splits:
## A.superior < 0.5 to the left, improve=3.262628, (0 missing)
## A.num.words < 117.5 to the right, improve=2.492701, (0 missing)
## A.num.chars < 999.5 to the right, improve=2.340345, (0 missing)
## A.num.words.unq < 73.5 to the right, improve=2.243030, (0 missing)
## T.advanc < 0.5 to the left, improve=2.181422, (0 missing)
## Surrogate splits:
## A.plus < 5.5 to the left, agree=0.972, adj=0.143, (0 split)
## A.fluorouracil < 1.5 to the left, agree=0.970, adj=0.071, (0 split)
##
## Node number 65: 19 observations
## predicted class=N expected loss=0.4210526 P(node) =0.01459293
## class counts: 11 8
## probabilities: 0.579 0.421
##
## Node number 66: 13 observations
## predicted class=N expected loss=0.1538462 P(node) =0.009984639
## class counts: 11 2
## probabilities: 0.846 0.154
##
## Node number 67: 12 observations
## predicted class=Y expected loss=0.25 P(node) =0.00921659
## class counts: 3 9
## probabilities: 0.250 0.750
##
## Node number 76: 192 observations, complexity param=0.01340326
## predicted class=N expected loss=0.3489583 P(node) =0.1474654
## class counts: 125 67
## probabilities: 0.651 0.349
## left son=152 (160 obs) right son=153 (32 obs)
## Primary splits:
## A.efficaci < 0.5 to the left, improve=7.252083, (0 missing)
## A.associ < 0.5 to the right, improve=5.683393, (0 missing)
## A.nodeposit < 0.5 to the left, improve=4.804801, (0 missing)
## T.versus < 0.5 to the left, improve=4.760051, (0 missing)
## A.oper < 0.5 to the left, improve=4.481559, (0 missing)
## Surrogate splits:
## A.firstlin < 1.5 to the left, agree=0.854, adj=0.125, (0 split)
## A.mgm2 < 2 to the left, agree=0.849, adj=0.094, (0 split)
## A.dose < 1.5 to the left, agree=0.849, adj=0.094, (0 split)
## A.safeti < 0.5 to the left, agree=0.849, adj=0.094, (0 split)
## A.docetaxel < 0.5 to the left, agree=0.844, adj=0.062, (0 split)
##
## Node number 77: 37 observations, complexity param=0.005244755
## predicted class=Y expected loss=0.2702703 P(node) =0.02841782
## class counts: 10 27
## probabilities: 0.270 0.730
## left son=154 (7 obs) right son=155 (30 obs)
## Primary splits:
## A.diseas < 1.5 to the right, improve=3.404118, (0 missing)
## A.therapi < 3.5 to the right, improve=2.568733, (0 missing)
## A.methotrex < 0.5 to the left, improve=2.002002, (0 missing)
## A.patient < 8.5 to the right, improve=1.935864, (0 missing)
## A.surviv < 1.5 to the right, improve=1.780930, (0 missing)
## Surrogate splits:
## A.patient < 9.5 to the right, agree=0.892, adj=0.429, (0 split)
## A.num.words.unq < 142.5 to the right, agree=0.892, adj=0.429, (0 split)
## A.site < 0.5 to the right, agree=0.892, adj=0.429, (0 split)
## A.num.chars < 2408.5 to the right, agree=0.892, adj=0.429, (0 split)
## A.num.words < 260 to the right, agree=0.892, adj=0.429, (0 split)
##
## Node number 78: 26 observations, complexity param=0.002622378
## predicted class=Y expected loss=0.3846154 P(node) =0.01996928
## class counts: 10 16
## probabilities: 0.385 0.615
## left son=156 (11 obs) right son=157 (15 obs)
## Primary splits:
## A.followup < 0.5 to the left, improve=2.416783, (0 missing)
## A.either < 0.5 to the left, improve=2.117216, (0 missing)
## A.use < 0.5 to the right, improve=2.082128, (0 missing)
## A.studi < 0.5 to the right, improve=1.568298, (0 missing)
## A.combin < 0.5 to the left, improve=1.557692, (0 missing)
## Surrogate splits:
## A.median < 0.5 to the left, agree=0.808, adj=0.545, (0 split)
## A.total < 0.5 to the right, agree=0.769, adj=0.455, (0 split)
## A.adjuv < 0.5 to the left, agree=0.769, adj=0.455, (0 split)
## A.month < 0.5 to the left, agree=0.731, adj=0.364, (0 split)
## A.cmf < 3.5 to the right, agree=0.731, adj=0.364, (0 split)
##
## Node number 79: 17 observations
## predicted class=Y expected loss=0 P(node) =0.01305684
## class counts: 0 17
## probabilities: 0.000 1.000
##
## Node number 94: 81 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.2345679 P(node) =0.06221198
## class counts: 19 62
## probabilities: 0.235 0.765
## left son=188 (10 obs) right son=189 (71 obs)
## Primary splits:
## A.metastas < 0.5 to the right, improve=3.046983, (0 missing)
## A.effect < 1.5 to the right, improve=2.838542, (0 missing)
## A.present < 0.5 to the right, improve=2.706283, (0 missing)
## A.metastat < 1.5 to the right, improve=2.384927, (0 missing)
## A.therapi < 0.5 to the right, improve=2.336742, (0 missing)
## Surrogate splits:
## A.month < 3.5 to the right, agree=0.926, adj=0.4, (0 split)
## A.object < 1.5 to the right, agree=0.901, adj=0.2, (0 split)
## A.site < 0.5 to the right, agree=0.901, adj=0.2, (0 split)
## A.studi < 3.5 to the right, agree=0.901, adj=0.2, (0 split)
## A.report < 1.5 to the right, agree=0.901, adj=0.2, (0 split)
##
## Node number 95: 47 observations
## predicted class=Y expected loss=0.0212766 P(node) =0.03609831
## class counts: 1 46
## probabilities: 0.021 0.979
##
## Node number 124: 68 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.2647059 P(node) =0.05222734
## class counts: 18 50
## probabilities: 0.265 0.735
## left son=248 (8 obs) right son=249 (60 obs)
## Primary splits:
## A.without < 0.5 to the right, improve=4.270588, (0 missing)
## A.receiv < 1.5 to the right, improve=3.351944, (0 missing)
## A.measur < 0.5 to the right, improve=3.351944, (0 missing)
## A.num.chars < 1103.5 to the right, improve=3.254689, (0 missing)
## A.num.words < 102 to the right, improve=3.254689, (0 missing)
## Surrogate splits:
## A.adjuv < 0.5 to the right, agree=0.912, adj=0.250, (0 split)
## A.increas < 1.5 to the right, agree=0.912, adj=0.250, (0 split)
## A.confirm < 0.5 to the right, agree=0.897, adj=0.125, (0 split)
## T.adjuv < 0.5 to the right, agree=0.897, adj=0.125, (0 split)
## A.incid < 0.5 to the right, agree=0.897, adj=0.125, (0 split)
##
## Node number 125: 61 observations
## predicted class=Y expected loss=0.04918033 P(node) =0.046851
## class counts: 3 58
## probabilities: 0.049 0.951
##
## Node number 128: 417 observations, complexity param=0.0006993007
## predicted class=N expected loss=0.08153477 P(node) =0.3202765
## class counts: 383 34
## probabilities: 0.918 0.082
## left son=256 (231 obs) right son=257 (186 obs)
## Primary splits:
## A.num.words < 124.5 to the right, improve=2.718588, (0 missing)
## T.advanc < 0.5 to the left, improve=2.619935, (0 missing)
## A.num.chars < 1410.5 to the right, improve=2.506831, (0 missing)
## A.num.words.unq < 35.5 to the right, improve=2.020412, (0 missing)
## A.among < 2.5 to the left, improve=1.714869, (0 missing)
## Surrogate splits:
## A.num.chars < 1423.5 to the right, agree=0.962, adj=0.914, (0 split)
## A.num.words.unq < 78.5 to the right, agree=0.906, adj=0.790, (0 split)
## A.breast < 0.5 to the right, agree=0.703, adj=0.333, (0 split)
## A.cancer < 0.5 to the right, agree=0.691, adj=0.306, (0 split)
## A.random < 0.5 to the right, agree=0.686, adj=0.296, (0 split)
##
## Node number 129: 14 observations
## predicted class=N expected loss=0.4285714 P(node) =0.01075269
## class counts: 8 6
## probabilities: 0.571 0.429
##
## Node number 152: 160 observations, complexity param=0.008741259
## predicted class=N expected loss=0.2875 P(node) =0.1228879
## class counts: 114 46
## probabilities: 0.713 0.288
## left son=304 (145 obs) right son=305 (15 obs)
## Primary splits:
## T.versus < 0.5 to the left, improve=4.759195, (0 missing)
## A.tamoxifen < 0.5 to the left, improve=4.567903, (0 missing)
## A.oper < 0.5 to the left, improve=3.648039, (0 missing)
## A.postop < 0.5 to the left, improve=3.630000, (0 missing)
## A.associ < 0.5 to the right, improve=3.310138, (0 missing)
## Surrogate splits:
## A.drug < 1.5 to the left, agree=0.919, adj=0.133, (0 split)
## A.respect < 2.5 to the left, agree=0.912, adj=0.067, (0 split)
## T.tamoxifen < 1.5 to the left, agree=0.912, adj=0.067, (0 split)
##
## Node number 153: 32 observations, complexity param=0.008741259
## predicted class=Y expected loss=0.34375 P(node) =0.02457757
## class counts: 11 21
## probabilities: 0.344 0.656
## left son=306 (11 obs) right son=307 (21 obs)
## Primary splits:
## A.data < 0.5 to the right, improve=4.931006, (0 missing)
## A.treatment < 1.5 to the left, improve=3.961310, (0 missing)
## A.random < 0.5 to the left, improve=2.611413, (0 missing)
## A.num.words < 129.5 to the left, improve=2.580357, (0 missing)
## A.well < 0.5 to the left, improve=2.520833, (0 missing)
## Surrogate splits:
## A.report < 0.5 to the right, agree=0.844, adj=0.545, (0 split)
## A.remain < 0.5 to the right, agree=0.781, adj=0.364, (0 split)
## A.avail < 0.5 to the right, agree=0.781, adj=0.364, (0 split)
## A.primari < 1.5 to the right, agree=0.750, adj=0.273, (0 split)
## A.includ < 0.5 to the right, agree=0.750, adj=0.273, (0 split)
##
## Node number 154: 7 observations
## predicted class=N expected loss=0.2857143 P(node) =0.005376344
## class counts: 5 2
## probabilities: 0.714 0.286
##
## Node number 155: 30 observations
## predicted class=Y expected loss=0.1666667 P(node) =0.02304147
## class counts: 5 25
## probabilities: 0.167 0.833
##
## Node number 156: 11 observations
## predicted class=N expected loss=0.3636364 P(node) =0.008448541
## class counts: 7 4
## probabilities: 0.636 0.364
##
## Node number 157: 15 observations
## predicted class=Y expected loss=0.2 P(node) =0.01152074
## class counts: 3 12
## probabilities: 0.200 0.800
##
## Node number 188: 10 observations
## predicted class=N expected loss=0.4 P(node) =0.007680492
## class counts: 6 4
## probabilities: 0.600 0.400
##
## Node number 189: 71 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.1830986 P(node) =0.05453149
## class counts: 13 58
## probabilities: 0.183 0.817
## left son=378 (7 obs) right son=379 (64 obs)
## Primary splits:
## A.present < 0.5 to the right, improve=2.342115, (0 missing)
## A.respons < 0.5 to the left, improve=2.141050, (0 missing)
## A.followup < 1.5 to the right, improve=1.810865, (0 missing)
## A.effect < 1.5 to the right, improve=1.521255, (0 missing)
## T.breast < 0.5 to the left, improve=1.407895, (0 missing)
## Surrogate splits:
## A.toxic < 4.5 to the right, agree=0.93, adj=0.286, (0 split)
##
## Node number 248: 8 observations
## predicted class=N expected loss=0.25 P(node) =0.006144393
## class counts: 6 2
## probabilities: 0.750 0.250
##
## Node number 249: 60 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.2 P(node) =0.04608295
## class counts: 12 48
## probabilities: 0.200 0.800
## left son=498 (32 obs) right son=499 (28 obs)
## Primary splits:
## T.studi < 0.5 to the left, improve=2.833929, (0 missing)
## A.receiv < 1.5 to the right, improve=2.186523, (0 missing)
## A.measur < 0.5 to the right, improve=2.186523, (0 missing)
## A.show < 0.5 to the right, improve=2.160000, (0 missing)
## T.iii < 0.5 to the right, improve=1.745455, (0 missing)
## Surrogate splits:
## T.trial < 0.5 to the right, agree=0.783, adj=0.536, (0 split)
## A.receiv < 0.5 to the right, agree=0.700, adj=0.357, (0 split)
## A.patient < 1.5 to the right, agree=0.683, adj=0.321, (0 split)
## A.num.chars < 1103.5 to the right, agree=0.683, adj=0.321, (0 split)
## A.num.words < 102 to the right, agree=0.683, adj=0.321, (0 split)
##
## Node number 256: 231 observations
## predicted class=N expected loss=0.03030303 P(node) =0.1774194
## class counts: 224 7
## probabilities: 0.970 0.030
##
## Node number 257: 186 observations, complexity param=0.0006993007
## predicted class=N expected loss=0.1451613 P(node) =0.1428571
## class counts: 159 27
## probabilities: 0.855 0.145
## left son=514 (165 obs) right son=515 (21 obs)
## Primary splits:
## T.advanc < 0.5 to the left, improve=2.632286, (0 missing)
## A.chemotherapi < 1.5 to the right, improve=1.753396, (0 missing)
## A.adjuv < 1.5 to the left, improve=1.694247, (0 missing)
## A.suggest < 0.5 to the left, improve=1.615519, (0 missing)
## A.overal < 0.5 to the left, improve=1.372654, (0 missing)
## Surrogate splits:
## T.compar < 0.5 to the left, agree=0.903, adj=0.143, (0 split)
## T.plus < 0.5 to the left, agree=0.892, adj=0.048, (0 split)
## A.remain < 0.5 to the left, agree=0.892, adj=0.048, (0 split)
##
## Node number 304: 145 observations, complexity param=0.008741259
## predicted class=N expected loss=0.2482759 P(node) =0.1113671
## class counts: 109 36
## probabilities: 0.752 0.248
## left son=608 (136 obs) right son=609 (9 obs)
## Primary splits:
## A.postop < 0.5 to the left, improve=5.380674, (0 missing)
## A.tamoxifen < 0.5 to the left, improve=3.492210, (0 missing)
## A.surviv < 1.5 to the left, improve=3.433229, (0 missing)
## A.oper < 0.5 to the left, improve=3.286740, (0 missing)
## A.plus < 1.5 to the left, improve=2.403335, (0 missing)
## Surrogate splits:
## A.seen < 1.5 to the left, agree=0.952, adj=0.222, (0 split)
## A.signific < 5.5 to the left, agree=0.945, adj=0.111, (0 split)
##
## Node number 305: 15 observations
## predicted class=Y expected loss=0.3333333 P(node) =0.01152074
## class counts: 5 10
## probabilities: 0.333 0.667
##
## Node number 306: 11 observations
## predicted class=N expected loss=0.2727273 P(node) =0.008448541
## class counts: 8 3
## probabilities: 0.727 0.273
##
## Node number 307: 21 observations
## predicted class=Y expected loss=0.1428571 P(node) =0.01612903
## class counts: 3 18
## probabilities: 0.143 0.857
##
## Node number 378: 7 observations
## predicted class=N expected loss=0.4285714 P(node) =0.005376344
## class counts: 4 3
## probabilities: 0.571 0.429
##
## Node number 379: 64 observations
## predicted class=Y expected loss=0.140625 P(node) =0.04915515
## class counts: 9 55
## probabilities: 0.141 0.859
##
## Node number 498: 32 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.34375 P(node) =0.02457757
## class counts: 11 21
## probabilities: 0.344 0.656
## left son=996 (25 obs) right son=997 (7 obs)
## Primary splits:
## A.cycl < 0.5 to the left, improve=2.117500, (0 missing)
## A.tamoxifen < 0.5 to the left, improve=2.117500, (0 missing)
## A.perform < 0.5 to the right, improve=1.687500, (0 missing)
## A.chemotherapi < 0.5 to the right, improve=1.660172, (0 missing)
## A.treatment < 1.5 to the left, improve=1.579200, (0 missing)
## Surrogate splits:
## A.diseasefre < 0.5 to the left, agree=0.906, adj=0.571, (0 split)
## A.cyclophosphamid < 0.5 to the left, agree=0.875, adj=0.429, (0 split)
## A.diseas < 2.5 to the left, agree=0.875, adj=0.429, (0 split)
## A.efficaci < 0.5 to the left, agree=0.875, adj=0.429, (0 split)
## A.event < 0.5 to the left, agree=0.875, adj=0.429, (0 split)
##
## Node number 499: 28 observations
## predicted class=Y expected loss=0.03571429 P(node) =0.02150538
## class counts: 1 27
## probabilities: 0.036 0.964
##
## Node number 514: 165 observations
## predicted class=N expected loss=0.1151515 P(node) =0.1267281
## class counts: 146 19
## probabilities: 0.885 0.115
##
## Node number 515: 21 observations, complexity param=0.0006993007
## predicted class=N expected loss=0.3809524 P(node) =0.01612903
## class counts: 13 8
## probabilities: 0.619 0.381
## left son=1030 (13 obs) right son=1031 (8 obs)
## Primary splits:
## A.num.words.unq < 49 to the left, improve=1.5393770, (0 missing)
## A.num.chars < 669.5 to the left, improve=1.5393770, (0 missing)
## A.num.words < 63.5 to the left, improve=1.5393770, (0 missing)
## A.respons < 0.5 to the left, improve=0.9603175, (0 missing)
## A.advanc < 0.5 to the left, improve=0.9603175, (0 missing)
## Surrogate splits:
## A.num.chars < 669.5 to the left, agree=1.000, adj=1.000, (0 split)
## A.num.words < 63.5 to the left, agree=1.000, adj=1.000, (0 split)
## A.patient < 0.5 to the left, agree=0.905, adj=0.750, (0 split)
## A.breast < 1.5 to the left, agree=0.905, adj=0.750, (0 split)
## A.activ < 0.5 to the left, agree=0.857, adj=0.625, (0 split)
##
## Node number 608: 136 observations, complexity param=0.002913753
## predicted class=N expected loss=0.2132353 P(node) =0.1044547
## class counts: 107 29
## probabilities: 0.787 0.213
## left son=1216 (123 obs) right son=1217 (13 obs)
## Primary splits:
## A.conduct < 0.5 to the left, improve=3.040733, (0 missing)
## T.clinic < 0.5 to the left, improve=2.641807, (0 missing)
## A.surviv < 1.5 to the left, improve=2.607059, (0 missing)
## A.tamoxifen < 0.5 to the left, improve=2.228782, (0 missing)
## T.advanc < 0.5 to the left, improve=2.165686, (0 missing)
## Surrogate splits:
## A.everi < 1.5 to the left, agree=0.919, adj=0.154, (0 split)
## A.alon < 1.5 to the left, agree=0.919, adj=0.154, (0 split)
## A.endpoint < 1.5 to the left, agree=0.912, adj=0.077, (0 split)
## A.treat < 2.5 to the left, agree=0.912, adj=0.077, (0 split)
## A.function. < 0.5 to the left, agree=0.912, adj=0.077, (0 split)
##
## Node number 609: 9 observations
## predicted class=Y expected loss=0.2222222 P(node) =0.006912442
## class counts: 2 7
## probabilities: 0.222 0.778
##
## Node number 996: 25 observations, complexity param=0.001748252
## predicted class=Y expected loss=0.44 P(node) =0.01920123
## class counts: 11 14
## probabilities: 0.440 0.560
## left son=1992 (9 obs) right son=1993 (16 obs)
## Primary splits:
## A.chemotherapi < 0.5 to the right, improve=3.208889, (0 missing)
## T.num.words < 10.5 to the right, improve=2.334706, (0 missing)
## T.num.words.unq < 10.5 to the right, improve=2.334706, (0 missing)
## T.num.chars < 104 to the right, improve=1.716825, (0 missing)
## A.receiv < 0.5 to the right, improve=1.666154, (0 missing)
## Surrogate splits:
## A.combin < 1.5 to the right, agree=0.80, adj=0.444, (0 split)
## A.regimen < 0.5 to the right, agree=0.80, adj=0.444, (0 split)
## A.median < 0.5 to the right, agree=0.76, adj=0.333, (0 split)
## A.surviv < 1.5 to the right, agree=0.76, adj=0.333, (0 split)
## A.patient < 3.5 to the right, agree=0.76, adj=0.333, (0 split)
##
## Node number 997: 7 observations
## predicted class=Y expected loss=0 P(node) =0.005376344
## class counts: 0 7
## probabilities: 0.000 1.000
##
## Node number 1030: 13 observations
## predicted class=N expected loss=0.2307692 P(node) =0.009984639
## class counts: 10 3
## probabilities: 0.769 0.231
##
## Node number 1031: 8 observations
## predicted class=Y expected loss=0.375 P(node) =0.006144393
## class counts: 3 5
## probabilities: 0.375 0.625
##
## Node number 1216: 123 observations, complexity param=0.002913753
## predicted class=N expected loss=0.1788618 P(node) =0.09447005
## class counts: 101 22
## probabilities: 0.821 0.179
## left son=2432 (102 obs) right son=2433 (21 obs)
## Primary splits:
## T.num.chars < 164.5 to the left, improve=3.158093, (0 missing)
## T.advanc < 0.5 to the left, improve=2.830081, (0 missing)
## T.clinic < 0.5 to the left, improve=2.755812, (0 missing)
## A.random < 2.5 to the left, improve=2.323088, (0 missing)
## A.sequenti < 0.5 to the left, improve=2.287717, (0 missing)
## Surrogate splits:
## T.num.words.unq < 15.5 to the left, agree=0.935, adj=0.619, (0 split)
## T.num.words < 15.5 to the left, agree=0.927, adj=0.571, (0 split)
## A.toler < 0.5 to the left, agree=0.854, adj=0.143, (0 split)
## A.number < 2.5 to the left, agree=0.854, adj=0.143, (0 split)
## T.cyclophosphamid < 0.5 to the left, agree=0.846, adj=0.095, (0 split)
##
## Node number 1217: 13 observations
## predicted class=Y expected loss=0.4615385 P(node) =0.009984639
## class counts: 6 7
## probabilities: 0.462 0.538
##
## Node number 1992: 9 observations
## predicted class=N expected loss=0.2222222 P(node) =0.006912442
## class counts: 7 2
## probabilities: 0.778 0.222
##
## Node number 1993: 16 observations
## predicted class=Y expected loss=0.25 P(node) =0.01228879
## class counts: 4 12
## probabilities: 0.250 0.750
##
## Node number 2432: 102 observations
## predicted class=N expected loss=0.127451 P(node) =0.07834101
## class counts: 89 13
## probabilities: 0.873 0.127
##
## Node number 2433: 21 observations, complexity param=0.002913753
## predicted class=N expected loss=0.4285714 P(node) =0.01612903
## class counts: 12 9
## probabilities: 0.571 0.429
## left son=4866 (9 obs) right son=4867 (12 obs)
## Primary splits:
## A.analysi < 0.5 to the right, improve=3.174603, (0 missing)
## A.num.words < 130 to the right, improve=2.670330, (0 missing)
## A.adjuv < 1.5 to the right, improve=1.994805, (0 missing)
## A.arm < 0.5 to the left, improve=1.714286, (0 missing)
## A.patient < 6.5 to the right, improve=1.714286, (0 missing)
## Surrogate splits:
## A.high < 0.5 to the right, agree=0.762, adj=0.444, (0 split)
## A.overal < 0.5 to the left, agree=0.714, adj=0.333, (0 split)
## A.assign < 0.5 to the right, agree=0.714, adj=0.333, (0 split)
## A.receiv < 1.5 to the right, agree=0.714, adj=0.333, (0 split)
## A.random < 1.5 to the right, agree=0.714, adj=0.333, (0 split)
##
## Node number 4866: 9 observations
## predicted class=N expected loss=0.1111111 P(node) =0.006912442
## class counts: 8 1
## probabilities: 0.889 0.111
##
## Node number 4867: 12 observations
## predicted class=Y expected loss=0.3333333 P(node) =0.00921659
## class counts: 4 8
## probabilities: 0.333 0.667
##
## n= 1302
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1302 572 N (0.56067588 0.43932412)
## 2) T.phase< 0.5 1005 322 N (0.67960199 0.32039801)
## 4) A.toxic< 0.5 811 196 N (0.75832306 0.24167694)
## 8) A.surviv< 0.5 484 66 N (0.86363636 0.13636364)
## 16) A.firstlin< 0.5 475 59 N (0.87578947 0.12421053)
## 32) A.mgm2< 0.5 450 48 N (0.89333333 0.10666667)
## 64) T.versus< 0.5 431 40 N (0.90719258 0.09280742)
## 128) A.superior< 0.5 417 34 N (0.91846523 0.08153477)
## 256) A.num.words>=124.5 231 7 N (0.96969697 0.03030303) *
## 257) A.num.words< 124.5 186 27 N (0.85483871 0.14516129)
## 514) T.advanc< 0.5 165 19 N (0.88484848 0.11515152) *
## 515) T.advanc>=0.5 21 8 N (0.61904762 0.38095238)
## 1030) A.num.words.unq< 49 13 3 N (0.76923077 0.23076923) *
## 1031) A.num.words.unq>=49 8 3 Y (0.37500000 0.62500000) *
## 129) A.superior>=0.5 14 6 N (0.57142857 0.42857143) *
## 65) T.versus>=0.5 19 8 N (0.57894737 0.42105263) *
## 33) A.mgm2>=0.5 25 11 N (0.56000000 0.44000000)
## 66) T.num.chars>=127.5 13 2 N (0.84615385 0.15384615) *
## 67) T.num.chars< 127.5 12 3 Y (0.25000000 0.75000000) *
## 17) A.firstlin>=0.5 9 2 Y (0.22222222 0.77777778) *
## 9) A.surviv>=0.5 327 130 N (0.60244648 0.39755352)
## 18) A.predict>=0.5 55 3 N (0.94545455 0.05454545) *
## 19) A.predict< 0.5 272 127 N (0.53308824 0.46691176)
## 38) A.cycl< 0.5 229 94 N (0.58951965 0.41048035)
## 76) A.durat< 0.5 192 67 N (0.65104167 0.34895833)
## 152) A.efficaci< 0.5 160 46 N (0.71250000 0.28750000)
## 304) T.versus< 0.5 145 36 N (0.75172414 0.24827586)
## 608) A.postop< 0.5 136 29 N (0.78676471 0.21323529)
## 1216) A.conduct< 0.5 123 22 N (0.82113821 0.17886179)
## 2432) T.num.chars< 164.5 102 13 N (0.87254902 0.12745098) *
## 2433) T.num.chars>=164.5 21 9 N (0.57142857 0.42857143)
## 4866) A.analysi>=0.5 9 1 N (0.88888889 0.11111111) *
## 4867) A.analysi< 0.5 12 4 Y (0.33333333 0.66666667) *
## 1217) A.conduct>=0.5 13 6 Y (0.46153846 0.53846154) *
## 609) A.postop>=0.5 9 2 Y (0.22222222 0.77777778) *
## 305) T.versus>=0.5 15 5 Y (0.33333333 0.66666667) *
## 153) A.efficaci>=0.5 32 11 Y (0.34375000 0.65625000)
## 306) A.data>=0.5 11 3 N (0.72727273 0.27272727) *
## 307) A.data< 0.5 21 3 Y (0.14285714 0.85714286) *
## 77) A.durat>=0.5 37 10 Y (0.27027027 0.72972973)
## 154) A.diseas>=1.5 7 2 N (0.71428571 0.28571429) *
## 155) A.diseas< 1.5 30 5 Y (0.16666667 0.83333333) *
## 39) A.cycl>=0.5 43 10 Y (0.23255814 0.76744186)
## 78) A.arm< 0.5 26 10 Y (0.38461538 0.61538462)
## 156) A.followup< 0.5 11 4 N (0.63636364 0.36363636) *
## 157) A.followup>=0.5 15 3 Y (0.20000000 0.80000000) *
## 79) A.arm>=0.5 17 0 Y (0.00000000 1.00000000) *
## 5) A.toxic>=0.5 194 68 Y (0.35051546 0.64948454)
## 10) A.reduc>=0.5 33 8 N (0.75757576 0.24242424)
## 20) A.clinic>=0.5 12 0 N (1.00000000 0.00000000) *
## 21) A.clinic< 0.5 21 8 N (0.61904762 0.38095238)
## 42) A.treat>=0.5 10 1 N (0.90000000 0.10000000) *
## 43) A.treat< 0.5 11 4 Y (0.36363636 0.63636364) *
## 11) A.reduc< 0.5 161 43 Y (0.26708075 0.73291925)
## 22) A.chang>=0.5 14 3 N (0.78571429 0.21428571) *
## 23) A.chang< 0.5 147 32 Y (0.21768707 0.78231293)
## 46) A.control>=0.5 19 7 N (0.63157895 0.36842105) *
## 47) A.control< 0.5 128 20 Y (0.15625000 0.84375000)
## 94) A.mgm2< 1.5 81 19 Y (0.23456790 0.76543210)
## 188) A.metastas>=0.5 10 4 N (0.60000000 0.40000000) *
## 189) A.metastas< 0.5 71 13 Y (0.18309859 0.81690141)
## 378) A.present>=0.5 7 3 N (0.57142857 0.42857143) *
## 379) A.present< 0.5 64 9 Y (0.14062500 0.85937500) *
## 95) A.mgm2>=1.5 47 1 Y (0.02127660 0.97872340) *
## 3) T.phase>=0.5 297 47 Y (0.15824916 0.84175084)
## 6) T.breast< 0.5 16 5 N (0.68750000 0.31250000) *
## 7) T.breast>=0.5 281 36 Y (0.12811388 0.87188612)
## 14) A.improv>=2.5 7 2 N (0.71428571 0.28571429) *
## 15) A.improv< 2.5 274 31 Y (0.11313869 0.88686131)
## 30) A.life>=0.5 12 6 N (0.50000000 0.50000000) *
## 31) A.life< 0.5 262 25 Y (0.09541985 0.90458015)
## 62) A.mgm2< 0.5 129 21 Y (0.16279070 0.83720930)
## 124) A.month< 0.5 68 18 Y (0.26470588 0.73529412)
## 248) A.without>=0.5 8 2 N (0.75000000 0.25000000) *
## 249) A.without< 0.5 60 12 Y (0.20000000 0.80000000)
## 498) T.studi< 0.5 32 11 Y (0.34375000 0.65625000)
## 996) A.cycl< 0.5 25 11 Y (0.44000000 0.56000000)
## 1992) A.chemotherapi>=0.5 9 2 N (0.77777778 0.22222222) *
## 1993) A.chemotherapi< 0.5 16 4 Y (0.25000000 0.75000000) *
## 997) A.cycl>=0.5 7 0 Y (0.00000000 1.00000000) *
## 499) T.studi>=0.5 28 1 Y (0.03571429 0.96428571) *
## 125) A.month>=0.5 61 3 Y (0.04918033 0.95081967) *
## 63) A.mgm2>=0.5 133 4 Y (0.03007519 0.96992481) *
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 288 10
## Y 442 562
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 288
## 2 Y 10
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 442
## 2 562
## Reference
## Prediction N Y
## N 551 46
## Y 179 526
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 551
## 2 Y 46
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 179
## 2 526
## Reference
## Prediction N Y
## N 603 63
## Y 127 509
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 603
## 2 Y 63
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 127
## 2 509
## Reference
## Prediction N Y
## N 633 79
## Y 97 493
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 633
## 2 Y 79
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 97
## 2 493
## Reference
## Prediction N Y
## N 662 100
## Y 68 472
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 662
## 2 Y 100
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 68
## 2 472
## Reference
## Prediction N Y
## N 674 113
## Y 56 459
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 674
## 2 Y 113
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 56
## 2 459
## Reference
## Prediction N Y
## N 690 143
## Y 40 429
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 690
## 2 Y 143
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 40
## 2 429
## Reference
## Prediction N Y
## N 704 190
## Y 26 382
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 704
## 2 Y 190
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 26
## 2 382
## Reference
## Prediction N Y
## N 721 288
## Y 9 284
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 721
## 2 Y 288
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 9
## 2 284
## Reference
## Prediction N Y
## N 730 548
## Y 0 24
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 730
## 2 Y 548
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 0
## 2 24
## threshold f.score
## 1 0.0 0.61045891
## 2 0.1 0.71319797
## 3 0.2 0.82380579
## 4 0.3 0.84271523
## 5 0.4 0.84853701
## 6 0.5 0.84892086
## 7 0.6 0.84452622
## 8 0.7 0.82420749
## 9 0.8 0.77959184
## 10 0.9 0.65664740
## 11 1.0 0.08053691
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 662
## 2 Y 100
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 68
## 2 472
## Reference
## Prediction N Y
## N 662 100
## Y 68 472
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 662
## 2 Y 100
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 68
## 2 472
## Prediction
## Reference N Y
## N 662 68
## Y 100 472
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.709677e-01 7.364830e-01 8.515346e-01 8.887079e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.192052e-129 1.677046e-02
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 128 15
## Y 185 230
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 128
## 2 Y 15
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 185
## 2 230
## Reference
## Prediction N Y
## N 219 44
## Y 94 201
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 219
## 2 Y 44
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 94
## 2 201
## Reference
## Prediction N Y
## N 231 60
## Y 82 185
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 231
## 2 Y 60
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 82
## 2 185
## Reference
## Prediction N Y
## N 242 74
## Y 71 171
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 242
## 2 Y 74
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 71
## 2 171
## Reference
## Prediction N Y
## N 257 81
## Y 56 164
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 257
## 2 Y 81
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 56
## 2 164
## Reference
## Prediction N Y
## N 262 84
## Y 51 161
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 262
## 2 Y 84
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 51
## 2 161
## Reference
## Prediction N Y
## N 270 102
## Y 43 143
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 270
## 2 Y 102
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 43
## 2 143
## Reference
## Prediction N Y
## N 285 121
## Y 28 124
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 285
## 2 Y 121
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 28
## 2 124
## Reference
## Prediction N Y
## N 302 150
## Y 11 95
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 302
## 2 Y 150
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 11
## 2 95
## Reference
## Prediction N Y
## N 311 237
## Y 2 8
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 311
## 2 Y 237
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 2
## 2 8
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6969697
## 3 0.2 0.7444444
## 4 0.3 0.7226562
## 5 0.4 0.7022587
## 6 0.5 0.7053763
## 7 0.6 0.7045952
## 8 0.7 0.6635731
## 9 0.8 0.6246851
## 10 0.9 0.5413105
## 11 1.0 0.0627451## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 219
## 2 Y 44
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 94
## 2 201
## Reference
## Prediction N Y
## N 219 44
## Y 94 201
## trial.fctr trial.fctr.predict.Conditional.X.cp.0.rpart.N
## 1 N 219
## 2 Y 44
## trial.fctr.predict.Conditional.X.cp.0.rpart.Y
## 1 94
## 2 201
## Prediction
## Reference N Y
## N 219 94
## Y 44 201
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.526882e-01 5.088091e-01 7.147118e-01 7.879644e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 3.903495e-21 3.030570e-05
## model_id model_method
## 1 Conditional.X.cp.0.rpart rpart
## feats
## 1 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 0 1.696 1.193
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9302483 0.5 0.8489209 0.8709677
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8515346 0.8887079 0.736483 0.8246854
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.7444444 0.7526882
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7147118 0.7879644 0.5088091
## [1] "fitting model: Conditional.X.rf"
## [1] " indep_vars: T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use"## Loading required package: randomForest
## randomForest 4.6-10
## Type rfNews() to see new features/changes/bug fixes.
## + : mtry= 2
## - : mtry= 2
## + : mtry=187
## - : mtry=187
## + : mtry=373
## - : mtry=373
## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 2 on full training set## Warning in myfit_mdl(model_id = paste0(model_id_pfx, ""), model_method =
## method, : model's bestTune found at an extreme of tuneGrid for parameter:
## mtry
## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 1302 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 2604 matrix numeric
## oob.times 1302 -none- numeric
## classes 2 -none- character
## importance 373 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 1302 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 373 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rf.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 116 4
## Y 614 568
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 116
## 2 Y 4
## trial.fctr.predict.Conditional.X.rf.Y
## 1 614
## 2 568
## Reference
## Prediction N Y
## N 526 8
## Y 204 564
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 526
## 2 Y 8
## trial.fctr.predict.Conditional.X.rf.Y
## 1 204
## 2 564
## Reference
## Prediction N Y
## N 681 11
## Y 49 561
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 681
## 2 Y 11
## trial.fctr.predict.Conditional.X.rf.Y
## 1 49
## 2 561
## Reference
## Prediction N Y
## N 720 14
## Y 10 558
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 720
## 2 Y 14
## trial.fctr.predict.Conditional.X.rf.Y
## 1 10
## 2 558
## Reference
## Prediction N Y
## N 729 23
## Y 1 549
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 729
## 2 Y 23
## trial.fctr.predict.Conditional.X.rf.Y
## 1 1
## 2 549
## Reference
## Prediction N Y
## N 730 60
## Y 0 512
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 730
## 2 Y 60
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 512
## Reference
## Prediction N Y
## N 730 132
## Y 0 440
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 730
## 2 Y 132
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 440
## Reference
## Prediction N Y
## N 730 311
## Y 0 261
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 730
## 2 Y 311
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 261
## Reference
## Prediction N Y
## N 730 542
## Y 0 30
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 730
## 2 Y 542
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 30
## Reference
## Prediction N Y
## N 730 572
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 730
## 2 Y 572
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.61045891
## 2 0.1 0.64766249
## 3 0.2 0.84179104
## 4 0.3 0.94923858
## 5 0.4 0.97894737
## 6 0.5 0.97860963
## 7 0.6 0.94464945
## 8 0.7 0.86956522
## 9 0.8 0.62665066
## 10 0.9 0.09966777
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 720
## 2 Y 14
## trial.fctr.predict.Conditional.X.rf.Y
## 1 10
## 2 558
## Reference
## Prediction N Y
## N 720 14
## Y 10 558
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 720
## 2 Y 14
## trial.fctr.predict.Conditional.X.rf.Y
## 1 10
## 2 558
## Prediction
## Reference N Y
## N 720 10
## Y 14 558
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.815668e-01 9.625543e-01 9.726964e-01 9.881547e-01 5.606759e-01
## AccuracyPValue McnemarPValue
## 1.433742e-279 5.402914e-01
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.rf.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 9 0
## Y 304 245
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 9
## 2 Y 0
## trial.fctr.predict.Conditional.X.rf.Y
## 1 304
## 2 245
## Reference
## Prediction N Y
## N 67 3
## Y 246 242
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 67
## 2 Y 3
## trial.fctr.predict.Conditional.X.rf.Y
## 1 246
## 2 242
## Reference
## Prediction N Y
## N 173 15
## Y 140 230
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 173
## 2 Y 15
## trial.fctr.predict.Conditional.X.rf.Y
## 1 140
## 2 230
## Reference
## Prediction N Y
## N 242 39
## Y 71 206
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 242
## 2 Y 39
## trial.fctr.predict.Conditional.X.rf.Y
## 1 71
## 2 206
## Reference
## Prediction N Y
## N 288 73
## Y 25 172
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 288
## 2 Y 73
## trial.fctr.predict.Conditional.X.rf.Y
## 1 25
## 2 172
## Reference
## Prediction N Y
## N 304 123
## Y 9 122
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 304
## 2 Y 123
## trial.fctr.predict.Conditional.X.rf.Y
## 1 9
## 2 122
## Reference
## Prediction N Y
## N 312 186
## Y 1 59
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 312
## 2 Y 186
## trial.fctr.predict.Conditional.X.rf.Y
## 1 1
## 2 59
## Reference
## Prediction N Y
## N 313 229
## Y 0 16
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 313
## 2 Y 229
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 16
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Conditional.X.rf.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6171285
## 3 0.2 0.6603001
## 4 0.3 0.7479675
## 5 0.4 0.7892720
## 6 0.5 0.7782805
## 7 0.6 0.6489362
## 8 0.7 0.3868852
## 9 0.8 0.1226054
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 242
## 2 Y 39
## trial.fctr.predict.Conditional.X.rf.Y
## 1 71
## 2 206
## Reference
## Prediction N Y
## N 242 39
## Y 71 206
## trial.fctr trial.fctr.predict.Conditional.X.rf.N
## 1 N 242
## 2 Y 39
## trial.fctr.predict.Conditional.X.rf.Y
## 1 71
## 2 206
## Prediction
## Reference N Y
## N 242 71
## Y 39 206
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.028674e-01 6.053900e-01 7.673939e-01 8.350960e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 1.887852e-33 3.119260e-03
## model_id model_method
## 1 Conditional.X.rf rf
## feats
## 1 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 136.335 25.034
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9876485 0.4 0.9789474 0.843318
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9726964 0.9881547 0.6755687 0.8972811
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.4 0.789272 0.8028674
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7673939 0.835096 0.60539
## [1] "fitting model: Conditional.X.no.rnorm.rf"
## [1] " indep_vars: T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use"
## + : mtry= 2
## - : mtry= 2
## + : mtry=187
## - : mtry=187
## + : mtry=372
## - : mtry=372
## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 187 on full training set
## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 1302 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 2604 matrix numeric
## oob.times 1302 -none- numeric
## classes 2 -none- character
## importance 372 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 1302 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 372 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 730
## 2 572
## Reference
## Prediction N Y
## N 459 0
## Y 271 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 459
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 271
## 2 572
## Reference
## Prediction N Y
## N 656 0
## Y 74 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 656
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 74
## 2 572
## Reference
## Prediction N Y
## N 720 0
## Y 10 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 720
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 10
## 2 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 572
## Reference
## Prediction N Y
## N 730 21
## Y 0 551
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 21
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 551
## Reference
## Prediction N Y
## N 730 101
## Y 0 471
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 101
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 471
## Reference
## Prediction N Y
## N 730 264
## Y 0 308
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 264
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 308
## Reference
## Prediction N Y
## N 730 566
## Y 0 6
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 566
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 6
## threshold f.score
## 1 0.0 0.61045891
## 2 0.1 0.80848057
## 3 0.2 0.93924466
## 4 0.3 0.99133449
## 5 0.4 1.00000000
## 6 0.5 1.00000000
## 7 0.6 1.00000000
## 8 0.7 0.98130009
## 9 0.8 0.90316395
## 10 0.9 0.70000000
## 11 1.0 0.02076125
## [1] "Classifier Probability Threshold: 0.6000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y NA
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 NA
## 2 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 730
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 572
## Prediction
## Reference N Y
## N 730 0
## Y 0 572
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 1.0000000 1.0000000 0.9971708 1.0000000 0.5606759
## AccuracyPValue McnemarPValue
## 0.0000000 NaN
## Reference
## Prediction N Y
## N 0 0
## Y 313 245
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 0
## 2 Y 0
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 313
## 2 245
## Reference
## Prediction N Y
## N 99 5
## Y 214 240
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 99
## 2 Y 5
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 214
## 2 240
## Reference
## Prediction N Y
## N 162 9
## Y 151 236
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 162
## 2 Y 9
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 151
## 2 236
## Reference
## Prediction N Y
## N 213 23
## Y 100 222
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 213
## 2 Y 23
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 100
## 2 222
## Reference
## Prediction N Y
## N 250 37
## Y 63 208
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 250
## 2 Y 37
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 63
## 2 208
## Reference
## Prediction N Y
## N 277 62
## Y 36 183
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 277
## 2 Y 62
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 36
## 2 183
## Reference
## Prediction N Y
## N 294 88
## Y 19 157
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 294
## 2 Y 88
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 19
## 2 157
## Reference
## Prediction N Y
## N 305 115
## Y 8 130
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 305
## 2 Y 115
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 8
## 2 130
## Reference
## Prediction N Y
## N 309 151
## Y 4 94
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 309
## 2 Y 151
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 4
## 2 94
## Reference
## Prediction N Y
## N 312 188
## Y 1 57
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 312
## 2 Y 188
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 1
## 2 57
## Reference
## Prediction N Y
## N 313 245
## Y 0 0
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 313
## 2 Y 245
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.6102117
## 2 0.1 0.6866953
## 3 0.2 0.7468354
## 4 0.3 0.7830688
## 5 0.4 0.8062016
## 6 0.5 0.7887931
## 7 0.6 0.7458432
## 8 0.7 0.6788512
## 9 0.8 0.5481050
## 10 0.9 0.3762376
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.OOB"
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 250
## 2 Y 37
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 63
## 2 208
## Reference
## Prediction N Y
## N 250 37
## Y 63 208
## trial.fctr trial.fctr.predict.Conditional.X.no.rnorm.rf.N
## 1 N 250
## 2 Y 37
## trial.fctr.predict.Conditional.X.no.rnorm.rf.Y
## 1 63
## 2 208
## Prediction
## Reference N Y
## N 250 63
## Y 37 208
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.207885e-01 6.403202e-01 7.863976e-01 8.517384e-01 5.609319e-01
## AccuracyPValue McnemarPValue
## 9.634624e-39 1.241933e-02
## model_id model_method
## 1 Conditional.X.no.rnorm.rf rf
## feats
## 1 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 141.822 33.202
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 1 0.6 1 0.8364055
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9971708 1 0.6664694 0.9037165
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.4 0.8062016 0.8207885
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7863976 0.8517384 0.6403202# User specified
# easier to exclude features
#model_id_pfx <- "";
# indep_vars_vctr <- setdiff(names(glb_trnent_df),
# union(union(glb_rsp_var, glb_exclude_vars_as_features),
# c("<feat1_name>", "<feat2_name>")))
# method <- ""
# easier to include features
#model_id_pfx <- ""; indep_vars_vctr <- c("<feat1_name>", "<feat1_name>"); method <- ""
# User specified bivariate models
# indep_vars_vctr_lst <- list()
# for (feat in setdiff(names(glb_trnent_df),
# union(glb_rsp_var, glb_exclude_vars_as_features)))
# indep_vars_vctr_lst[["feat"]] <- feat
# User specified combinatorial models
# indep_vars_vctr_lst <- list()
# combn_mtrx <- combn(c("<feat1_name>", "<feat2_name>", "<featn_name>"),
# <num_feats_to_choose>)
# for (combn_ix in 1:ncol(combn_mtrx))
# #print(combn_mtrx[, combn_ix])
# indep_vars_vctr_lst[[combn_ix]] <- combn_mtrx[, combn_ix]
# template for myfit_mdl
# rf is hard-coded in caret to recognize only Accuracy / Kappa evaluation metrics
# only for OOB in trainControl ?
# ret_lst <- myfit_mdl_fn(model_id=paste0(model_id_pfx, ""), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_trnent_df, OOB_df=glb_newent_df,
# n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df,
# model_loss_mtrx=glb_model_metric_terms,
# model_summaryFunction=glb_model_metric_smmry,
# model_metric=glb_model_metric,
# model_metric_maximize=glb_model_metric_maximize)
# Simplify a model
# fit_df <- glb_trnent_df; glb_mdl <- step(<complex>_mdl)
# Non-caret models
# rpart_area_mdl <- rpart(reformulate("Area", response=glb_rsp_var),
# data=glb_trnent_df, #method="class",
# control=rpart.control(cp=0.12),
# parms=list(loss=glb_model_metric_terms))
# print("rpart_sel_wlm_mdl"); prp(rpart_sel_wlm_mdl)
#
print(glb_models_df)## model_id model_method
## 1 MFO.myMFO_classfr myMFO_classfr
## 2 Random.myrandom_classfr myrandom_classfr
## 3 Max.cor.Y.cv.0.rpart rpart
## 4 Max.cor.Y.cv.0.cp.0.rpart rpart
## 5 Max.cor.Y.rpart rpart
## 6 Max.cor.Y.glm glm
## 7 Interact.High.cor.Y.glm glm
## 8 Low.cor.X.glm glm
## 9 Conditional.X.glm glm
## 10 Conditional.X.rpart rpart
## 11 Conditional.X.cp.0.rpart rpart
## 12 Conditional.X.rf rf
## 13 Conditional.X.no.rnorm.rf rf
## feats
## 1 .rnorm
## 2 .rnorm
## 3 T.phase
## 4 T.phase
## 5 T.phase
## 6 T.phase
## 7 T.phase, T.phase:T.cancer, T.phase:A.nausea, T.phase:T.num.words, T.phase:T.num.words.unq, T.phase:A.point, T.phase:A.breast, T.phase:A.life
## 8 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, A.two, A.object, A.rang, A.paclitaxel, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 9 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 10 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 11 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 12 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 13 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 0 0.380 0.002
## 2 0 0.243 0.001
## 3 0 0.601 0.020
## 4 0 0.473 0.017
## 5 3 0.984 0.018
## 6 1 0.873 0.021
## 7 1 0.949 0.046
## 8 1 18.582 5.480
## 9 1 19.673 6.063
## 10 3 7.181 1.182
## 11 0 1.696 1.193
## 12 3 136.335 25.034
## 13 3 141.822 33.202
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5000000 0.5 0.0000000 0.5606759
## 2 0.5097711 0.4 0.6104589 0.4393241
## 3 0.5000000 0.5 0.0000000 0.5606759
## 4 0.6863397 0.3 0.6104589 0.4393241
## 5 0.6863397 0.3 0.6104589 0.7165899
## 6 0.6863397 0.3 0.6104589 0.7165899
## 7 0.6936201 0.3 0.6107848 0.7150538
## 8 0.9374940 0.9 0.9307282 0.7058372
## 9 0.9418910 0.9 0.9342561 0.7058372
## 10 0.7631694 0.6 0.7074318 0.7588326
## 11 0.9302483 0.5 0.8489209 0.8709677
## 12 0.9876485 0.4 0.9789474 0.8433180
## 13 1.0000000 0.6 1.0000000 0.8364055
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.5332221 0.5878552 0.0000000 0.5000000
## 2 0.4121448 0.4667779 0.0000000 0.4970333
## 3 0.5332221 0.5878552 0.0000000 0.5000000
## 4 0.4121448 0.4667779 0.0000000 0.6739975
## 5 0.4121448 0.4667779 0.3930612 0.6739975
## 6 0.4121448 0.4667779 0.3930612 0.6739975
## 7 0.4129061 0.4675493 0.3883401 0.6848667
## 8 0.9257933 0.9523607 0.4082901 0.7201213
## 9 0.9274808 0.9537375 0.4074552 0.7239356
## 10 0.7370177 0.7840707 0.5017505 0.7635587
## 11 0.8515346 0.8887079 0.7364830 0.8246854
## 12 0.9726964 0.9881547 0.6755687 0.8972811
## 13 0.9971708 1.0000000 0.6664694 0.9037165
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0.0000000 0.5609319
## 2 0.4 0.6102117 0.4390681
## 3 0.5 0.0000000 0.5609319
## 4 0.3 0.6102117 0.4390681
## 5 0.3 0.6102117 0.4390681
## 6 0.3 0.6102117 0.4390681
## 7 0.3 0.6125000 0.4444444
## 8 0.9 0.6832298 0.7258065
## 9 0.9 0.6910569 0.7275986
## 10 0.6 0.7056180 0.7652330
## 11 0.2 0.7444444 0.7526882
## 12 0.4 0.7892720 0.8028674
## 13 0.4 0.8062016 0.8207885
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.5186280 0.6025874 0.000000000
## 2 0.3974126 0.4813720 0.000000000
## 3 0.5186280 0.6025874 0.000000000
## 4 0.3974126 0.4813720 0.000000000
## 5 0.3974126 0.4813720 0.000000000
## 6 0.3974126 0.4813720 0.000000000
## 7 0.4027083 0.4867718 0.008426483
## 8 0.6867647 0.7624378 0.441613144
## 9 0.6886224 0.7641450 0.447474366
## 10 0.7278160 0.7998142 0.513685051
## 11 0.7147118 0.7879644 0.508809083
## 12 0.7673939 0.8350960 0.605390045
## 13 0.7863976 0.8517384 0.640320231
## max.AccuracySD.fit max.KappaSD.fit min.aic.fit
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 0.02887088 0.063422125 NA
## 6 0.02887088 0.063422125 1524.042
## 7 0.02815715 0.061614822 1504.884
## 8 0.01463330 0.030934654 6356.810
## 9 0.00133030 0.002711058 6226.635
## 10 0.01350107 0.028641541 NA
## 11 NA NA NA
## 12 NA NA NA
## 13 NA NA NAglb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.models",
chunk_step_major=glb_script_df[nrow(glb_script_df), "chunk_step_major"],
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed9 fit.models 5 1 89.153
## elapsed10 fit.models 5 2 426.511if (!is.null(glb_model_metric_smmry)) {
stats_df <- glb_models_df[, "model_id", FALSE]
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_trnent_df, glb_rsp_var,
glb_rsp_var_out, model_id, "fit",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_newent_df, glb_rsp_var,
glb_rsp_var_out, model_id, "OOB",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
# tmp_models_df <- orderBy(~model_id, glb_models_df)
# rownames(tmp_models_df) <- seq(1, nrow(tmp_models_df))
# all.equal(subset(tmp_models_df[, names(stats_df)], model_id != "Random.myrandom_classfr"),
# subset(stats_df, model_id != "Random.myrandom_classfr"))
# print(subset(tmp_models_df[, names(stats_df)], model_id != "Random.myrandom_classfr")[, c("model_id", "max.Accuracy.fit")])
# print(subset(stats_df, model_id != "Random.myrandom_classfr")[, c("model_id", "max.Accuracy.fit")])
print("Merging following data into glb_models_df:")
print(stats_mrg_df <- stats_df[, c(1, grep(glb_model_metric, names(stats_df)))])
print(tmp_models_df <- orderBy(~model_id, glb_models_df[, c("model_id", grep(glb_model_metric, names(stats_df), value=TRUE))]))
tmp2_models_df <- glb_models_df[, c("model_id", setdiff(names(glb_models_df), grep(glb_model_metric, names(stats_df), value=TRUE)))]
tmp3_models_df <- merge(tmp2_models_df, stats_mrg_df, all.x=TRUE, sort=FALSE)
print(tmp3_models_df)
print(names(tmp3_models_df))
print(glb_models_df <- subset(tmp3_models_df, select=-model_id.1))
}
plt_models_df <- glb_models_df[, -grep("SD|Upper|Lower", names(glb_models_df))]
for (var in grep("^min.", names(plt_models_df), value=TRUE)) {
plt_models_df[, sub("min.", "inv.", var)] <-
#ifelse(all(is.na(tmp <- plt_models_df[, var])), NA, 1.0 / tmp)
1.0 / plt_models_df[, var]
plt_models_df <- plt_models_df[ , -grep(var, names(plt_models_df))]
}
print(plt_models_df)## model_id model_method
## 1 MFO.myMFO_classfr myMFO_classfr
## 2 Random.myrandom_classfr myrandom_classfr
## 3 Max.cor.Y.cv.0.rpart rpart
## 4 Max.cor.Y.cv.0.cp.0.rpart rpart
## 5 Max.cor.Y.rpart rpart
## 6 Max.cor.Y.glm glm
## 7 Interact.High.cor.Y.glm glm
## 8 Low.cor.X.glm glm
## 9 Conditional.X.glm glm
## 10 Conditional.X.rpart rpart
## 11 Conditional.X.cp.0.rpart rpart
## 12 Conditional.X.rf rf
## 13 Conditional.X.no.rnorm.rf rf
## feats
## 1 .rnorm
## 2 .rnorm
## 3 T.phase
## 4 T.phase
## 5 T.phase
## 6 T.phase
## 7 T.phase, T.phase:T.cancer, T.phase:A.nausea, T.phase:T.num.words, T.phase:T.num.words.unq, T.phase:A.point, T.phase:A.breast, T.phase:A.life
## 8 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, A.two, A.object, A.rang, A.paclitaxel, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 9 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 10 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 11 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 12 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, .rnorm, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## 13 T.phase, A.toxic, A.mgm2, T.metastat, A.everi, A.median, A.rate, T.studi, T.versus, T.advanc, A.respons, A.metastat, A.progress, A.advanc, A.neutropenia, A.partial, A.toler, A.combin, A.firstlin, A.regimen, T.breast, A.cyclophosphamid, T.docetaxel, T.cyclophosphamid, A.overal, A.week, A.month, A.surviv, T.combin, A.day, A.arm, A.docetaxel, T.cancer, T.iii, T.compar, A.durat, T.plus, A.given, A.epirubicin, A.mbc, A.previous, A.X100, A.four, A.cycl, A.methotrex, T.group, A.phase, A.diseas, A.efficaci, A.X500, A.grade, A.patient, A.occur, A.hundr, A.schedul, T.doxorubicin, A.either, A.dose, A.activ, A.progressionfre, A.safeti, A.doxorubicin, A.seen, A.elig, T.trial, A.hematolog, A.plus, A.vomit, A.respect, A.fluorouracil, A.X5fluorouracil, A.superior, A.enter, T.num.chars, A.two, A.object, A.rang, A.paclitaxel, T.num.words, A.three, T.random, A.prior, A.cmf, A.sequenti, A.oral, A.complet, A.diseasefre, A.achiev, A.similar, T.randomis, A.six, A.secondari, A.well, T.num.words.unq, A.event, A.anthracyclin, A.death, A.longer, A.eight, A.one, A.enrol, A.term, A.administ, A.failur, A.common, A.frequent, A.assign, A.nodeposit, A.respond, A.endpoint, A.stabl, A.confid, A.prolong, A.continu, A.daili, A.experienc, A.follow, A.treatment, A.end, A.time, A.receiv, A.dfs, A.primari, A.interv, A.compar, A.total, T.treatment, A.intraven, T.tamoxifen, T.result, A.cours, A.observ, A.advers, A.evalu, A.versus, A.pretreat, A.point, T.postmenopaus, A.includ, A.seven, A.nausea, A.five, A.better, A.multicent, A.oper, A.stage, A.alon, A.less, A.random, A.although, A.addit, A.agent, A.due, A.andor, A.lymph, A.followup, A.postmenopaus, A.involv, A.side, A.standard, A.receptorposit, A.conduct, A.neoadjuv, A.major, A.administr, A.limit, A.consist, A.differ, A.num.words.unq, A.singl, A.hazard, A.sever, A.result, A.randomis, A.appear, A.treat, A.profil, A.year, A.accord, T.patient, A.site, A.improv, A.tamoxifen, A.ratio, A.evid, A.iii, A.conclus, A.demonstr, A.statist, A.among, A.relaps, A.human, A.without, A.surgeri, A.drug, A.infus, T.therapi, A.higher, A.purpos, A.premenopaus, A.num.chars, A.background, A.benefit, A.growth, T.chemotherapi, A.confirm, A.inhibitor, A.characterist, A.whether, A.popul, A.initi, A.group, A.mastectomi, A.least, A.consid, A.node, A.axillari, A.num.words, A.caus, A.endocrin, A.estim, A.new, A.postop, A.distant, A.regard, A.requir, A.defin, A.trend, A.subgroup, A.receptor, A.greater, A.therapi, A.studi, A.howev, A.support, T.adjuv, A.number, A.develop, A.comparison, A.prospect, A.regress, A.per, A.age, A.general, A.earli, A.X001, A.base, A.carcinoma, A.method, A.patholog, A.set, A.her2, A.status, A.start, A.incid, A.determin, A.posit, A.progesteron, A.aromatas, A.local, A.main, A.present, A.aim, A.possibl, A.perform, A.larg, A.shown, A.hormon, A.period, A.reduct, T.clinic, A.second, A.breast, A.estrogen, A.cancer, A.tissu, A.adjuv, A.case, A.find, A.obtain, A.suggest, A.within, A.remain, A.design, A.X0001, A.potenti, A.import, A.detect, A.women, A.recurr, A.investig, A.chemotherapi, A.histolog, A.lower, T.earli, A.select, A.first, A.signific, A.low, A.mean, T.women, A.radiotherapi, A.clinic, A.also, A.relat, A.life, A.size, A.effect, A.outcom, A.wherea, A.may, A.indic, A.system, A.test, A.metastas, A.type, A.function., A.analysi, A.avail, A.negat, A.analys, A.X005, A.report, A.show, A.provid, A.tumor, A.qualiti, A.can, A.doubleblind, A.tumour, A.serum, A.factor, A.analyz, A.proport, A.valu, A.sampl, A.inform, A.assess, A.need, A.independ, A.correl, A.increas, A.level, A.placebo, A.multivari, A.model, A.trial, A.high, A.score, A.express, A.found, A.particip, A.control, A.examin, A.prognost, A.marker, A.reduc, A.cell, A.measur, A.chang, T.respons, A.baselin, A.identifi, A.decreas, A.bone, A.prevent, A.data, A.associ, A.predict, T.effect, A.risk, A.use
## max.nTuningRuns max.auc.fit opt.prob.threshold.fit max.f.score.fit
## 1 0 0.5000000 0.5 0.0000000
## 2 0 0.5097711 0.4 0.6104589
## 3 0 0.5000000 0.5 0.0000000
## 4 0 0.6863397 0.3 0.6104589
## 5 3 0.6863397 0.3 0.6104589
## 6 1 0.6863397 0.3 0.6104589
## 7 1 0.6936201 0.3 0.6107848
## 8 1 0.9374940 0.9 0.9307282
## 9 1 0.9418910 0.9 0.9342561
## 10 3 0.7631694 0.6 0.7074318
## 11 0 0.9302483 0.5 0.8489209
## 12 3 0.9876485 0.4 0.9789474
## 13 3 1.0000000 0.6 1.0000000
## max.Accuracy.fit max.Kappa.fit max.auc.OOB opt.prob.threshold.OOB
## 1 0.5606759 0.0000000 0.5000000 0.5
## 2 0.4393241 0.0000000 0.4970333 0.4
## 3 0.5606759 0.0000000 0.5000000 0.5
## 4 0.4393241 0.0000000 0.6739975 0.3
## 5 0.7165899 0.3930612 0.6739975 0.3
## 6 0.7165899 0.3930612 0.6739975 0.3
## 7 0.7150538 0.3883401 0.6848667 0.3
## 8 0.7058372 0.4082901 0.7201213 0.9
## 9 0.7058372 0.4074552 0.7239356 0.9
## 10 0.7588326 0.5017505 0.7635587 0.6
## 11 0.8709677 0.7364830 0.8246854 0.2
## 12 0.8433180 0.6755687 0.8972811 0.4
## 13 0.8364055 0.6664694 0.9037165 0.4
## max.f.score.OOB max.Accuracy.OOB max.Kappa.OOB
## 1 0.0000000 0.5609319 0.000000000
## 2 0.6102117 0.4390681 0.000000000
## 3 0.0000000 0.5609319 0.000000000
## 4 0.6102117 0.4390681 0.000000000
## 5 0.6102117 0.4390681 0.000000000
## 6 0.6102117 0.4390681 0.000000000
## 7 0.6125000 0.4444444 0.008426483
## 8 0.6832298 0.7258065 0.441613144
## 9 0.6910569 0.7275986 0.447474366
## 10 0.7056180 0.7652330 0.513685051
## 11 0.7444444 0.7526882 0.508809083
## 12 0.7892720 0.8028674 0.605390045
## 13 0.8062016 0.8207885 0.640320231
## inv.elapsedtime.everything inv.elapsedtime.final inv.aic.fit
## 1 2.631578947 5.000000e+02 NA
## 2 4.115226337 1.000000e+03 NA
## 3 1.663893511 5.000000e+01 NA
## 4 2.114164905 5.882353e+01 NA
## 5 1.016260163 5.555556e+01 NA
## 6 1.145475372 4.761905e+01 0.0006561497
## 7 1.053740780 2.173913e+01 0.0006645030
## 8 0.053815520 1.824818e-01 0.0001573116
## 9 0.050831088 1.649349e-01 0.0001606004
## 10 0.139256371 8.460237e-01 NA
## 11 0.589622642 8.382230e-01 NA
## 12 0.007334874 3.994567e-02 NA
## 13 0.007051092 3.011867e-02 NAprint(myplot_radar(radar_inp_df=plt_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 13. Consider specifying shapes manually. if you must have them.## Warning: Removed 6 rows containing missing values (geom_path).## Warning: Removed 104 rows containing missing values (geom_point).## Warning: Removed 9 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 13. Consider specifying shapes manually. if you must have them.
# print(myplot_radar(radar_inp_df=subset(plt_models_df,
# !(model_id %in% grep("random|MFO", plt_models_df$model_id, value=TRUE)))))
# Compute CI for <metric>SD
glb_models_df <- mutate(glb_models_df,
max.df = ifelse(max.nTuningRuns > 1, max.nTuningRuns - 1, NA),
min.sd2ci.scaler = ifelse(is.na(max.df), NA, qt(0.975, max.df)))
for (var in grep("SD", names(glb_models_df), value=TRUE)) {
# Does CI alredy exist ?
var_components <- unlist(strsplit(var, "SD"))
varActul <- paste0(var_components[1], var_components[2])
varUpper <- paste0(var_components[1], "Upper", var_components[2])
varLower <- paste0(var_components[1], "Lower", var_components[2])
if (varUpper %in% names(glb_models_df)) {
warning(varUpper, " already exists in glb_models_df")
# Assuming Lower also exists
next
}
print(sprintf("var:%s", var))
# CI is dependent on sample size in t distribution; df=n-1
glb_models_df[, varUpper] <- glb_models_df[, varActul] +
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
glb_models_df[, varLower] <- glb_models_df[, varActul] -
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
}## Warning: max.AccuracyUpper.fit already exists in glb_models_df## [1] "var:max.KappaSD.fit"# Plot metrics with CI
plt_models_df <- glb_models_df[, "model_id", FALSE]
pltCI_models_df <- glb_models_df[, "model_id", FALSE]
for (var in grep("Upper", names(glb_models_df), value=TRUE)) {
var_components <- unlist(strsplit(var, "Upper"))
col_name <- unlist(paste(var_components, collapse=""))
plt_models_df[, col_name] <- glb_models_df[, col_name]
for (name in paste0(var_components[1], c("Upper", "Lower"), var_components[2]))
pltCI_models_df[, name] <- glb_models_df[, name]
}
build_statsCI_data <- function(plt_models_df) {
mltd_models_df <- melt(plt_models_df, id.vars="model_id")
mltd_models_df$data <- sapply(1:nrow(mltd_models_df),
function(row_ix) tail(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]), "[.]")), 1))
mltd_models_df$label <- sapply(1:nrow(mltd_models_df),
function(row_ix) head(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]), paste0(".", mltd_models_df[row_ix, "data"]))), 1))
#print(mltd_models_df)
return(mltd_models_df)
}
mltd_models_df <- build_statsCI_data(plt_models_df)
mltdCI_models_df <- melt(pltCI_models_df, id.vars="model_id")
for (row_ix in 1:nrow(mltdCI_models_df)) {
for (type in c("Upper", "Lower")) {
if (length(var_components <- unlist(strsplit(
as.character(mltdCI_models_df[row_ix, "variable"]), type))) > 1) {
#print(sprintf("row_ix:%d; type:%s; ", row_ix, type))
mltdCI_models_df[row_ix, "label"] <- var_components[1]
mltdCI_models_df[row_ix, "data"] <- unlist(strsplit(var_components[2], "[.]"))[2]
mltdCI_models_df[row_ix, "type"] <- type
break
}
}
}
#print(mltdCI_models_df)
# castCI_models_df <- dcast(mltdCI_models_df, value ~ type, fun.aggregate=sum)
# print(castCI_models_df)
wideCI_models_df <- reshape(subset(mltdCI_models_df, select=-variable),
timevar="type",
idvar=setdiff(names(mltdCI_models_df), c("type", "value", "variable")),
direction="wide")
#print(wideCI_models_df)
mrgdCI_models_df <- merge(wideCI_models_df, mltd_models_df, all.x=TRUE)
#print(mrgdCI_models_df)
# Merge stats back in if CIs don't exist
goback_vars <- c()
for (var in unique(mltd_models_df$label)) {
for (type in unique(mltd_models_df$data)) {
var_type <- paste0(var, ".", type)
# if this data is already present, next
if (var_type %in% unique(paste(mltd_models_df$label, mltd_models_df$data, sep=".")))
next
#print(sprintf("var_type:%s", var_type))
goback_vars <- c(goback_vars, var_type)
}
}
if (length(goback_vars) > 0) {
mltd_goback_df <- build_statsCI_data(glb_models_df[, c("model_id", goback_vars)])
mltd_models_df <- rbind(mltd_models_df, mltd_goback_df)
}
mltd_models_df <- merge(mltd_models_df, glb_models_df[, c("model_id", "model_method")], all.x=TRUE)
png(paste0(glb_out_pfx, "models_bar.png"), width=480*3, height=480*2)
print(gp <- myplot_bar(mltd_models_df, "model_id", "value", colorcol_name="model_method") +
geom_errorbar(data=mrgdCI_models_df,
mapping=aes(x=model_id, ymax=value.Upper, ymin=value.Lower), width=0.5) +
facet_grid(label ~ data, scales="free") +
theme(axis.text.x = element_text(angle = 90,vjust = 0.5)))
dev.off()## quartz_off_screen
## 2print(gp)
# used for console inspection
model_evl_terms <- c(NULL)
for (metric in glb_model_evl_criteria)
model_evl_terms <- c(model_evl_terms,
ifelse(length(grep("max", metric)) > 0, "-", "+"), metric)
model_sel_frmla <- as.formula(paste(c("~ ", model_evl_terms), collapse=" "))
print(tmp_models_df <- orderBy(model_sel_frmla, glb_models_df)[, c("model_id", glb_model_evl_criteria)])## model_id max.Accuracy.OOB max.Kappa.OOB min.aic.fit
## 13 Conditional.X.no.rnorm.rf 0.8207885 0.640320231 NA
## 12 Conditional.X.rf 0.8028674 0.605390045 NA
## 10 Conditional.X.rpart 0.7652330 0.513685051 NA
## 11 Conditional.X.cp.0.rpart 0.7526882 0.508809083 NA
## 9 Conditional.X.glm 0.7275986 0.447474366 6226.635
## 8 Low.cor.X.glm 0.7258065 0.441613144 6356.810
## 1 MFO.myMFO_classfr 0.5609319 0.000000000 NA
## 3 Max.cor.Y.cv.0.rpart 0.5609319 0.000000000 NA
## 7 Interact.High.cor.Y.glm 0.4444444 0.008426483 1504.884
## 6 Max.cor.Y.glm 0.4390681 0.000000000 1524.042
## 2 Random.myrandom_classfr 0.4390681 0.000000000 NA
## 4 Max.cor.Y.cv.0.cp.0.rpart 0.4390681 0.000000000 NA
## 5 Max.cor.Y.rpart 0.4390681 0.000000000 NAprint(myplot_radar(radar_inp_df=tmp_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 13. Consider specifying shapes manually. if you must have them.## Warning: Removed 6 rows containing missing values (geom_path).## Warning: Removed 27 rows containing missing values (geom_point).## Warning: Removed 9 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 13. Consider specifying shapes manually. if you must have them.
print("Metrics used for model selection:"); print(model_sel_frmla)## [1] "Metrics used for model selection:"## ~-max.Accuracy.OOB - max.Kappa.OOB + min.aic.fitprint(sprintf("Best model id: %s", tmp_models_df[1, "model_id"]))## [1] "Best model id: Conditional.X.no.rnorm.rf"if (is.null(glb_sel_mdl_id))
{ glb_sel_mdl_id <- tmp_models_df[1, "model_id"] } else
print(sprintf("User specified selection: %s", glb_sel_mdl_id))
myprint_mdl(glb_sel_mdl <- glb_models_lst[[glb_sel_mdl_id]])
## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 1302 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 2604 matrix numeric
## oob.times 1302 -none- numeric
## classes 2 -none- character
## importance 372 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 1302 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 372 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character## [1] TRUEreplay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"model.selected")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.data.training.all",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed10 fit.models 5 2 426.511
## elapsed11 fit.data.training.all 6 0 440.471Step 6: fit.data.training.all
if (!is.null(glb_fin_mdl_id) && (glb_fin_mdl_id %in% names(glb_models_lst))) {
warning("Final model same as user selected model")
glb_fin_mdl <- glb_sel_mdl
} else {
print(mdl_feats_df <- myextract_mdl_feats(sel_mdl=glb_sel_mdl, entity_df=glb_trnent_df))
if ((model_method <- glb_sel_mdl$method) == "custom")
# get actual method from the model_id
model_method <- tail(unlist(strsplit(glb_sel_mdl_id, "[.]")), 1)
tune_finmdl_df <- NULL
if (nrow(glb_sel_mdl$bestTune) > 0) {
for (param in names(glb_sel_mdl$bestTune)) {
#print(sprintf("param: %s", param))
tune_finmdl_df <- rbind(tune_finmdl_df,
data.frame(parameter=param,
min=glb_sel_mdl$bestTune[1, param],
max=glb_sel_mdl$bestTune[1, param],
by=1)) # by val does not matter
}
}
# Sync with parameters in mydsutils.R
ret_lst <- myfit_mdl(model_id="Final", model_method=model_method,
indep_vars_vctr=mdl_feats_df$id, model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=NULL,
n_cv_folds=glb_n_cv_folds, tune_models_df=tune_finmdl_df,
# Automate from here
# Issues if glb_sel_mdl$method == "rf" b/c trainControl is "oob"; not "cv"
model_loss_mtrx=glb_model_metric_terms,
model_summaryFunction=glb_sel_mdl$control$summaryFunction,
model_metric=glb_sel_mdl$metric,
model_metric_maximize=glb_sel_mdl$maximize)
glb_fin_mdl <- glb_models_lst[[length(glb_models_lst)]]
glb_fin_mdl_id <- glb_models_df[length(glb_models_lst), "model_id"]
}## importance id fit.feat
## T.phase 1.000000e+02 T.phase TRUE
## A.toxic 6.823672e+01 A.toxic TRUE
## A.mgm2 5.099683e+01 A.mgm2 TRUE
## A.surviv 2.782371e+01 A.surviv TRUE
## T.versus 1.906868e+01 T.versus TRUE
## A.respons 1.770692e+01 A.respons TRUE
## A.median 1.695478e+01 A.median TRUE
## A.rate 1.185187e+01 A.rate TRUE
## T.num.chars 1.171642e+01 T.num.chars TRUE
## A.cyclophosphamid 1.091006e+01 A.cyclophosphamid TRUE
## T.advanc 1.081725e+01 T.advanc TRUE
## A.num.chars 9.671727e+00 A.num.chars TRUE
## T.breast 9.668191e+00 T.breast TRUE
## A.num.words 8.650528e+00 A.num.words TRUE
## A.num.words.unq 8.466277e+00 A.num.words.unq TRUE
## A.predict 7.541403e+00 A.predict TRUE
## T.num.words.unq 6.632736e+00 T.num.words.unq TRUE
## T.num.words 6.098768e+00 T.num.words TRUE
## T.cyclophosphamid 5.780144e+00 T.cyclophosphamid TRUE
## A.use 5.693311e+00 A.use TRUE
## A.everi 5.657050e+00 A.everi TRUE
## T.chemotherapi 5.652411e+00 T.chemotherapi TRUE
## A.risk 5.069597e+00 A.risk TRUE
## A.cell 5.015032e+00 A.cell TRUE
## A.tamoxifen 4.879619e+00 A.tamoxifen TRUE
## T.studi 4.341576e+00 T.studi TRUE
## A.month 4.290409e+00 A.month TRUE
## A.reduc 4.218943e+00 A.reduc TRUE
## A.chemotherapi 4.151334e+00 A.chemotherapi TRUE
## A.firstlin 3.964696e+00 A.firstlin TRUE
## A.placebo 3.775989e+00 A.placebo TRUE
## A.patient 3.755779e+00 A.patient TRUE
## A.prevent 3.582255e+00 A.prevent TRUE
## A.control 3.482417e+00 A.control TRUE
## A.trial 3.429586e+00 A.trial TRUE
## T.respons 3.348781e+00 T.respons TRUE
## A.day 3.144477e+00 A.day TRUE
## A.arm 3.108747e+00 A.arm TRUE
## A.durat 3.100409e+00 A.durat TRUE
## A.receiv 3.011959e+00 A.receiv TRUE
## A.dose 2.966946e+00 A.dose TRUE
## A.breast 2.938524e+00 A.breast TRUE
## A.treatment 2.770097e+00 A.treatment TRUE
## A.increas 2.727862e+00 A.increas TRUE
## A.data 2.711346e+00 A.data TRUE
## A.bone 2.703806e+00 A.bone TRUE
## A.either 2.697808e+00 A.either TRUE
## T.tamoxifen 2.689161e+00 T.tamoxifen TRUE
## A.fluorouracil 2.619404e+00 A.fluorouracil TRUE
## A.year 2.614689e+00 A.year TRUE
## A.toler 2.563663e+00 A.toler TRUE
## A.alon 2.561795e+00 A.alon TRUE
## A.associ 2.544094e+00 A.associ TRUE
## A.chang 2.510186e+00 A.chang TRUE
## A.methotrex 2.506642e+00 A.methotrex TRUE
## A.radiotherapi 2.490141e+00 A.radiotherapi TRUE
## A.metastas 2.473860e+00 A.metastas TRUE
## A.cycl 2.341962e+00 A.cycl TRUE
## A.combin 2.317690e+00 A.combin TRUE
## A.efficaci 2.311089e+00 A.efficaci TRUE
## A.postop 2.275071e+00 A.postop TRUE
## A.compar 2.264635e+00 A.compar TRUE
## A.overal 2.241060e+00 A.overal TRUE
## A.cmf 2.204722e+00 A.cmf TRUE
## A.regimen 2.198447e+00 A.regimen TRUE
## A.advanc 2.196970e+00 A.advanc TRUE
## A.week 2.188930e+00 A.week TRUE
## A.seen 2.158254e+00 A.seen TRUE
## A.women 2.146172e+00 A.women TRUE
## A.improv 2.120010e+00 A.improv TRUE
## A.cours 2.113454e+00 A.cours TRUE
## A.random 2.101144e+00 A.random TRUE
## A.baselin 2.069444e+00 A.baselin TRUE
## T.cancer 2.057827e+00 T.cancer TRUE
## T.metastat 2.057544e+00 T.metastat TRUE
## A.term 2.003160e+00 A.term TRUE
## T.adjuv 2.001591e+00 T.adjuv TRUE
## A.given 1.978028e+00 A.given TRUE
## A.among 1.921062e+00 A.among TRUE
## A.two 1.920949e+00 A.two TRUE
## A.cancer 1.915703e+00 A.cancer TRUE
## A.group 1.878725e+00 A.group TRUE
## A.diseasefre 1.872521e+00 A.diseasefre TRUE
## A.high 1.820940e+00 A.high TRUE
## A.therapi 1.817669e+00 A.therapi TRUE
## A.differ 1.814999e+00 A.differ TRUE
## A.progress 1.812191e+00 A.progress TRUE
## A.clinic 1.798282e+00 A.clinic TRUE
## A.oral 1.784579e+00 A.oral TRUE
## A.daili 1.781438e+00 A.daili TRUE
## A.total 1.773837e+00 A.total TRUE
## T.compar 1.768398e+00 T.compar TRUE
## T.combin 1.750313e+00 T.combin TRUE
## T.treatment 1.741618e+00 T.treatment TRUE
## A.analysi 1.714107e+00 A.analysi TRUE
## A.identifi 1.686683e+00 A.identifi TRUE
## A.signific 1.685476e+00 A.signific TRUE
## T.patient 1.683084e+00 T.patient TRUE
## A.effect 1.665246e+00 A.effect TRUE
## A.postmenopaus 1.648046e+00 A.postmenopaus TRUE
## A.relaps 1.596956e+00 A.relaps TRUE
## A.studi 1.570836e+00 A.studi TRUE
## T.women 1.543079e+00 T.women TRUE
## A.superior 1.519245e+00 A.superior TRUE
## A.measur 1.517929e+00 A.measur TRUE
## A.prognost 1.517210e+00 A.prognost TRUE
## A.factor 1.502419e+00 A.factor TRUE
## A.time 1.501231e+00 A.time TRUE
## A.surgeri 1.499885e+00 A.surgeri TRUE
## A.evid 1.477139e+00 A.evid TRUE
## A.intraven 1.470883e+00 A.intraven TRUE
## A.stage 1.469443e+00 A.stage TRUE
## T.clinic 1.465538e+00 T.clinic TRUE
## T.result 1.446680e+00 T.result TRUE
## T.trial 1.442182e+00 T.trial TRUE
## A.level 1.429940e+00 A.level TRUE
## A.pretreat 1.427429e+00 A.pretreat TRUE
## A.X100 1.413110e+00 A.X100 TRUE
## A.assess 1.407650e+00 A.assess TRUE
## A.recurr 1.388128e+00 A.recurr TRUE
## A.life 1.381300e+00 A.life TRUE
## A.system 1.379889e+00 A.system TRUE
## A.event 1.374947e+00 A.event TRUE
## A.dfs 1.357746e+00 A.dfs TRUE
## A.obtain 1.341322e+00 A.obtain TRUE
## T.effect 1.333944e+00 T.effect TRUE
## A.without 1.323480e+00 A.without TRUE
## T.group 1.308616e+00 T.group TRUE
## A.randomis 1.298374e+00 A.randomis TRUE
## T.randomis 1.297974e+00 T.randomis TRUE
## A.nodeposit 1.283348e+00 A.nodeposit TRUE
## A.previous 1.274568e+00 A.previous TRUE
## A.adjuv 1.238004e+00 A.adjuv TRUE
## A.followup 1.231614e+00 A.followup TRUE
## A.statist 1.225913e+00 A.statist TRUE
## A.primari 1.225184e+00 A.primari TRUE
## A.first 1.194686e+00 A.first TRUE
## A.treat 1.181920e+00 A.treat TRUE
## A.assign 1.175217e+00 A.assign TRUE
## A.develop 1.167349e+00 A.develop TRUE
## A.benefit 1.165625e+00 A.benefit TRUE
## A.activ 1.152097e+00 A.activ TRUE
## A.show 1.149203e+00 A.show TRUE
## A.carcinoma 1.136952e+00 A.carcinoma TRUE
## A.analyz 1.120803e+00 A.analyz TRUE
## A.result 1.119868e+00 A.result TRUE
## A.doxorubicin 1.119564e+00 A.doxorubicin TRUE
## A.respond 1.103757e+00 A.respond TRUE
## A.agent 1.094820e+00 A.agent TRUE
## A.plus 1.073349e+00 A.plus TRUE
## A.administ 1.067449e+00 A.administ TRUE
## A.versus 1.067337e+00 A.versus TRUE
## A.follow 1.044566e+00 A.follow TRUE
## A.achiev 1.040276e+00 A.achiev TRUE
## A.failur 1.027160e+00 A.failur TRUE
## A.metastat 1.023877e+00 A.metastat TRUE
## A.standard 1.022125e+00 A.standard TRUE
## T.plus 1.020776e+00 T.plus TRUE
## A.decreas 1.005570e+00 A.decreas TRUE
## A.iii 1.005358e+00 A.iii TRUE
## A.consid 9.957225e-01 A.consid TRUE
## A.administr 9.943735e-01 A.administr TRUE
## A.complet 9.922000e-01 A.complet TRUE
## A.tumor 9.909364e-01 A.tumor TRUE
## A.express 9.904643e-01 A.express TRUE
## A.hundr 9.898229e-01 A.hundr TRUE
## A.suggest 9.824333e-01 A.suggest TRUE
## A.estrogen 9.782543e-01 A.estrogen TRUE
## A.site 9.743949e-01 A.site TRUE
## A.respect 9.727954e-01 A.respect TRUE
## A.endpoint 9.570354e-01 A.endpoint TRUE
## A.three 9.379247e-01 A.three TRUE
## A.paclitaxel 9.117324e-01 A.paclitaxel TRUE
## A.observ 9.108305e-01 A.observ TRUE
## A.wherea 9.106555e-01 A.wherea TRUE
## A.sequenti 9.077838e-01 A.sequenti TRUE
## A.oper 8.907183e-01 A.oper TRUE
## A.schedul 8.898603e-01 A.schedul TRUE
## A.side 8.824742e-01 A.side TRUE
## A.evalu 8.719362e-01 A.evalu TRUE
## A.reduct 8.564211e-01 A.reduct TRUE
## A.analys 8.541659e-01 A.analys TRUE
## A.epirubicin 8.502762e-01 A.epirubicin TRUE
## A.earli 8.458413e-01 A.earli TRUE
## T.earli 8.392187e-01 T.earli TRUE
## A.X5fluorouracil 8.277524e-01 A.X5fluorouracil TRUE
## A.axillari 8.131148e-01 A.axillari TRUE
## A.anthracyclin 8.107474e-01 A.anthracyclin TRUE
## A.conduct 8.100448e-01 A.conduct TRUE
## A.advers 8.098925e-01 A.advers TRUE
## A.phase 8.071224e-01 A.phase TRUE
## A.receptor 8.059120e-01 A.receptor TRUE
## A.status 8.042197e-01 A.status TRUE
## A.lymph 7.955726e-01 A.lymph TRUE
## A.initi 7.892223e-01 A.initi TRUE
## A.receptorposit 7.853589e-01 A.receptorposit TRUE
## A.provid 7.851682e-01 A.provid TRUE
## A.outcom 7.640508e-01 A.outcom TRUE
## T.postmenopaus 7.623659e-01 T.postmenopaus TRUE
## A.perform 7.589352e-01 A.perform TRUE
## A.multivari 7.586241e-01 A.multivari TRUE
## A.safeti 7.505672e-01 A.safeti TRUE
## T.therapi 7.402377e-01 T.therapi TRUE
## A.qualiti 7.300010e-01 A.qualiti TRUE
## A.hormon 7.271770e-01 A.hormon TRUE
## A.rang 7.218339e-01 A.rang TRUE
## A.caus 7.138734e-01 A.caus TRUE
## A.confid 7.050701e-01 A.confid TRUE
## A.serum 7.038422e-01 A.serum TRUE
## A.appear 6.895936e-01 A.appear TRUE
## A.docetaxel 6.891831e-01 A.docetaxel TRUE
## A.four 6.831992e-01 A.four TRUE
## A.diseas 6.765381e-01 A.diseas TRUE
## A.model 6.750900e-01 A.model TRUE
## A.conclus 6.713914e-01 A.conclus TRUE
## A.prolong 6.665404e-01 A.prolong TRUE
## A.independ 6.577949e-01 A.independ TRUE
## A.one 6.546509e-01 A.one TRUE
## A.continu 6.463054e-01 A.continu TRUE
## A.may 6.440620e-01 A.may TRUE
## T.random 6.358129e-01 T.random TRUE
## A.object 6.255294e-01 A.object TRUE
## A.found 6.216466e-01 A.found TRUE
## A.period 6.215341e-01 A.period TRUE
## A.report 6.168596e-01 A.report TRUE
## A.enter 6.133765e-01 A.enter TRUE
## A.within 6.119043e-01 A.within TRUE
## A.low 6.067003e-01 A.low TRUE
## A.progressionfre 6.023040e-01 A.progressionfre TRUE
## A.similar 6.003831e-01 A.similar TRUE
## A.although 6.000208e-01 A.although TRUE
## A.indic 5.999529e-01 A.indic TRUE
## A.X001 5.976633e-01 A.X001 TRUE
## A.occur 5.935623e-01 A.occur TRUE
## A.longer 5.922324e-01 A.longer TRUE
## A.larg 5.920315e-01 A.larg TRUE
## A.addit 5.906567e-01 A.addit TRUE
## A.endocrin 5.881546e-01 A.endocrin TRUE
## A.correl 5.869908e-01 A.correl TRUE
## A.inform 5.839487e-01 A.inform TRUE
## A.partial 5.808240e-01 A.partial TRUE
## A.involv 5.797309e-01 A.involv TRUE
## A.function. 5.741905e-01 A.function. TRUE
## A.drug 5.660788e-01 A.drug TRUE
## A.examin 5.652573e-01 A.examin TRUE
## A.incid 5.624113e-01 A.incid TRUE
## A.method 5.543479e-01 A.method TRUE
## A.select 5.499096e-01 A.select TRUE
## A.determin 5.468575e-01 A.determin TRUE
## A.age 5.425819e-01 A.age TRUE
## A.X500 5.395976e-01 A.X500 TRUE
## A.import 5.328980e-01 A.import TRUE
## A.nausea 5.318831e-01 A.nausea TRUE
## A.posit 5.270544e-01 A.posit TRUE
## A.also 5.243409e-01 A.also TRUE
## A.premenopaus 5.207826e-01 A.premenopaus TRUE
## A.avail 5.201772e-01 A.avail TRUE
## A.progesteron 5.171904e-01 A.progesteron TRUE
## A.less 5.123328e-01 A.less TRUE
## A.mastectomi 5.108349e-01 A.mastectomi TRUE
## A.whether 5.072933e-01 A.whether TRUE
## A.better 4.893985e-01 A.better TRUE
## A.end 4.891119e-01 A.end TRUE
## A.tissu 4.888862e-01 A.tissu TRUE
## A.neoadjuv 4.886702e-01 A.neoadjuv TRUE
## A.interv 4.776809e-01 A.interv TRUE
## A.investig 4.776775e-01 A.investig TRUE
## A.can 4.764178e-01 A.can TRUE
## A.lower 4.754371e-01 A.lower TRUE
## A.node 4.699306e-01 A.node TRUE
## A.prospect 4.640852e-01 A.prospect TRUE
## A.X005 4.580343e-01 A.X005 TRUE
## A.purpos 4.574152e-01 A.purpos TRUE
## A.second 4.524330e-01 A.second TRUE
## A.trend 4.469123e-01 A.trend TRUE
## A.aim 4.428552e-01 A.aim TRUE
## A.distant 4.409518e-01 A.distant TRUE
## A.ratio 4.396401e-01 A.ratio TRUE
## A.stabl 4.381738e-01 A.stabl TRUE
## A.mean 4.351955e-01 A.mean TRUE
## A.popul 4.350026e-01 A.popul TRUE
## A.infus 4.338135e-01 A.infus TRUE
## A.shown 4.328210e-01 A.shown TRUE
## A.includ 4.312279e-01 A.includ TRUE
## A.possibl 4.307860e-01 A.possibl TRUE
## A.common 4.254047e-01 A.common TRUE
## A.hazard 4.236241e-01 A.hazard TRUE
## A.present 4.166014e-01 A.present TRUE
## A.number 4.163898e-01 A.number TRUE
## T.doxorubicin 4.154180e-01 T.doxorubicin TRUE
## A.proport 4.016981e-01 A.proport TRUE
## A.consist 3.964571e-01 A.consist TRUE
## A.well 3.946550e-01 A.well TRUE
## A.remain 3.923603e-01 A.remain TRUE
## A.mbc 3.906713e-01 A.mbc TRUE
## A.doubleblind 3.830866e-01 A.doubleblind TRUE
## A.accord 3.774967e-01 A.accord TRUE
## A.local 3.703015e-01 A.local TRUE
## A.higher 3.635092e-01 A.higher TRUE
## A.start 3.606558e-01 A.start TRUE
## A.relat 3.587580e-01 A.relat TRUE
## A.tumour 3.579266e-01 A.tumour TRUE
## A.design 3.558210e-01 A.design TRUE
## A.secondari 3.550964e-01 A.secondari TRUE
## A.confirm 3.548021e-01 A.confirm TRUE
## A.demonstr 3.504067e-01 A.demonstr TRUE
## A.estim 3.466403e-01 A.estim TRUE
## A.negat 3.442517e-01 A.negat TRUE
## A.main 3.434569e-01 A.main TRUE
## A.sampl 3.326324e-01 A.sampl TRUE
## A.six 3.301388e-01 A.six TRUE
## A.base 3.210605e-01 A.base TRUE
## A.sever 3.201116e-01 A.sever TRUE
## A.per 3.141204e-01 A.per TRUE
## A.find 3.112677e-01 A.find TRUE
## A.growth 3.094469e-01 A.growth TRUE
## T.iii 3.079152e-01 T.iii TRUE
## A.enrol 3.035677e-01 A.enrol TRUE
## A.elig 3.004993e-01 A.elig TRUE
## A.neutropenia 3.004672e-01 A.neutropenia TRUE
## A.valu 2.994850e-01 A.valu TRUE
## A.howev 2.915030e-01 A.howev TRUE
## A.vomit 2.827938e-01 A.vomit TRUE
## A.least 2.800623e-01 A.least TRUE
## A.X0001 2.783991e-01 A.X0001 TRUE
## A.greater 2.763786e-01 A.greater TRUE
## A.death 2.715267e-01 A.death TRUE
## T.docetaxel 2.692344e-01 T.docetaxel TRUE
## A.due 2.678806e-01 A.due TRUE
## A.grade 2.671869e-01 A.grade TRUE
## A.case 2.572374e-01 A.case TRUE
## A.marker 2.467158e-01 A.marker TRUE
## A.major 2.422363e-01 A.major TRUE
## A.experienc 2.405727e-01 A.experienc TRUE
## A.frequent 2.343557e-01 A.frequent TRUE
## A.need 2.327583e-01 A.need TRUE
## A.patholog 2.254064e-01 A.patholog TRUE
## A.singl 2.186042e-01 A.singl TRUE
## A.comparison 2.162860e-01 A.comparison TRUE
## A.support 2.160740e-01 A.support TRUE
## A.prior 2.155730e-01 A.prior TRUE
## A.eight 2.105145e-01 A.eight TRUE
## A.particip 2.091339e-01 A.particip TRUE
## A.histolog 2.057285e-01 A.histolog TRUE
## A.potenti 2.003699e-01 A.potenti TRUE
## A.subgroup 1.961642e-01 A.subgroup TRUE
## A.inhibitor 1.938515e-01 A.inhibitor TRUE
## A.detect 1.916176e-01 A.detect TRUE
## A.test 1.908182e-01 A.test TRUE
## A.background 1.891662e-01 A.background TRUE
## A.point 1.861108e-01 A.point TRUE
## A.her2 1.813718e-01 A.her2 TRUE
## A.size 1.764043e-01 A.size TRUE
## A.seven 1.759985e-01 A.seven TRUE
## A.five 1.579276e-01 A.five TRUE
## A.requir 1.578153e-01 A.requir TRUE
## A.new 1.540105e-01 A.new TRUE
## A.type 1.361213e-01 A.type TRUE
## A.andor 1.249572e-01 A.andor TRUE
## A.set 1.207486e-01 A.set TRUE
## A.regard 1.187030e-01 A.regard TRUE
## A.score 9.929894e-02 A.score TRUE
## A.general 9.016826e-02 A.general TRUE
## A.regress 7.975996e-02 A.regress TRUE
## A.limit 6.564507e-02 A.limit TRUE
## A.defin 6.404772e-02 A.defin TRUE
## A.aromatas 5.890749e-02 A.aromatas TRUE
## A.multicent 4.523932e-02 A.multicent TRUE
## A.characterist 4.109205e-02 A.characterist TRUE
## A.human 3.739331e-02 A.human TRUE
## A.profil 5.315485e-03 A.profil TRUE
## A.hematolog 0.000000e+00 A.hematolog TRUE
## [1] "fitting model: Final.rf"
## [1] " indep_vars: T.phase, A.toxic, A.mgm2, A.surviv, T.versus, A.respons, A.median, A.rate, T.num.chars, A.cyclophosphamid, T.advanc, A.num.chars, T.breast, A.num.words, A.num.words.unq, A.predict, T.num.words.unq, T.num.words, T.cyclophosphamid, A.use, A.everi, T.chemotherapi, A.risk, A.cell, A.tamoxifen, T.studi, A.month, A.reduc, A.chemotherapi, A.firstlin, A.placebo, A.patient, A.prevent, A.control, A.trial, T.respons, A.day, A.arm, A.durat, A.receiv, A.dose, A.breast, A.treatment, A.increas, A.data, A.bone, A.either, T.tamoxifen, A.fluorouracil, A.year, A.toler, A.alon, A.associ, A.chang, A.methotrex, A.radiotherapi, A.metastas, A.cycl, A.combin, A.efficaci, A.postop, A.compar, A.overal, A.cmf, A.regimen, A.advanc, A.week, A.seen, A.women, A.improv, A.cours, A.random, A.baselin, T.cancer, T.metastat, A.term, T.adjuv, A.given, A.among, A.two, A.cancer, A.group, A.diseasefre, A.high, A.therapi, A.differ, A.progress, A.clinic, A.oral, A.daili, A.total, T.compar, T.combin, T.treatment, A.analysi, A.identifi, A.signific, T.patient, A.effect, A.postmenopaus, A.relaps, A.studi, T.women, A.superior, A.measur, A.prognost, A.factor, A.time, A.surgeri, A.evid, A.intraven, A.stage, T.clinic, T.result, T.trial, A.level, A.pretreat, A.X100, A.assess, A.recurr, A.life, A.system, A.event, A.dfs, A.obtain, T.effect, A.without, T.group, A.randomis, T.randomis, A.nodeposit, A.previous, A.adjuv, A.followup, A.statist, A.primari, A.first, A.treat, A.assign, A.develop, A.benefit, A.activ, A.show, A.carcinoma, A.analyz, A.result, A.doxorubicin, A.respond, A.agent, A.plus, A.administ, A.versus, A.follow, A.achiev, A.failur, A.metastat, A.standard, T.plus, A.decreas, A.iii, A.consid, A.administr, A.complet, A.tumor, A.express, A.hundr, A.suggest, A.estrogen, A.site, A.respect, A.endpoint, A.three, A.paclitaxel, A.observ, A.wherea, A.sequenti, A.oper, A.schedul, A.side, A.evalu, A.reduct, A.analys, A.epirubicin, A.earli, T.earli, A.X5fluorouracil, A.axillari, A.anthracyclin, A.conduct, A.advers, A.phase, A.receptor, A.status, A.lymph, A.initi, A.receptorposit, A.provid, A.outcom, T.postmenopaus, A.perform, A.multivari, A.safeti, T.therapi, A.qualiti, A.hormon, A.rang, A.caus, A.confid, A.serum, A.appear, A.docetaxel, A.four, A.diseas, A.model, A.conclus, A.prolong, A.independ, A.one, A.continu, A.may, T.random, A.object, A.found, A.period, A.report, A.enter, A.within, A.low, A.progressionfre, A.similar, A.although, A.indic, A.X001, A.occur, A.longer, A.larg, A.addit, A.endocrin, A.correl, A.inform, A.partial, A.involv, A.function., A.drug, A.examin, A.incid, A.method, A.select, A.determin, A.age, A.X500, A.import, A.nausea, A.posit, A.also, A.premenopaus, A.avail, A.progesteron, A.less, A.mastectomi, A.whether, A.better, A.end, A.tissu, A.neoadjuv, A.interv, A.investig, A.can, A.lower, A.node, A.prospect, A.X005, A.purpos, A.second, A.trend, A.aim, A.distant, A.ratio, A.stabl, A.mean, A.popul, A.infus, A.shown, A.includ, A.possibl, A.common, A.hazard, A.present, A.number, T.doxorubicin, A.proport, A.consist, A.well, A.remain, A.mbc, A.doubleblind, A.accord, A.local, A.higher, A.start, A.relat, A.tumour, A.design, A.secondari, A.confirm, A.demonstr, A.estim, A.negat, A.main, A.sampl, A.six, A.base, A.sever, A.per, A.find, A.growth, T.iii, A.enrol, A.elig, A.neutropenia, A.valu, A.howev, A.vomit, A.least, A.X0001, A.greater, A.death, T.docetaxel, A.due, A.grade, A.case, A.marker, A.major, A.experienc, A.frequent, A.need, A.patholog, A.singl, A.comparison, A.support, A.prior, A.eight, A.particip, A.histolog, A.potenti, A.subgroup, A.inhibitor, A.detect, A.test, A.background, A.point, A.her2, A.size, A.seven, A.five, A.requir, A.new, A.type, A.andor, A.set, A.regard, A.score, A.general, A.regress, A.limit, A.defin, A.aromatas, A.multicent, A.characterist, A.human, A.profil, A.hematolog"
## + : mtry=187
## - : mtry=187
## Aggregating results
## Fitting final model on full training set
## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 1302 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 2604 matrix numeric
## oob.times 1302 -none- numeric
## classes 2 -none- character
## importance 372 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 1302 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 372 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## Reference
## Prediction N Y
## N 0 0
## Y 730 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 0 730
## 2 Y 0 572
## Reference
## Prediction N Y
## N 465 0
## Y 265 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 465 265
## 2 Y 0 572
## Reference
## Prediction N Y
## N 655 0
## Y 75 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 655 75
## 2 Y 0 572
## Reference
## Prediction N Y
## N 717 0
## Y 13 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 717 13
## 2 Y 0 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 0 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 0 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 0 572
## Reference
## Prediction N Y
## N 730 24
## Y 0 548
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 24 548
## Reference
## Prediction N Y
## N 730 101
## Y 0 471
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 101 471
## Reference
## Prediction N Y
## N 730 258
## Y 0 314
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 258 314
## Reference
## Prediction N Y
## N 730 566
## Y 0 6
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 566 6
## threshold f.score
## 1 0.0 0.61045891
## 2 0.1 0.81192335
## 3 0.2 0.93847416
## 4 0.3 0.98876404
## 5 0.4 1.00000000
## 6 0.5 1.00000000
## 7 0.6 1.00000000
## 8 0.7 0.97857143
## 9 0.8 0.90316395
## 10 0.9 0.70880361
## 11 1.0 0.02076125## [1] "Classifier Probability Threshold: 0.6000 to maximize f.score.fit"
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 NA
## 2 Y NA 572
## Reference
## Prediction N Y
## N 730 0
## Y 0 572
## trial.fctr trial.fctr.predict.Final.rf.N trial.fctr.predict.Final.rf.Y
## 1 N 730 0
## 2 Y 0 572
## Prediction
## Reference N Y
## N 730 0
## Y 0 572
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 1.0000000 1.0000000 0.9971708 1.0000000 0.5606759
## AccuracyPValue McnemarPValue
## 0.0000000 NaN## Warning in mypredict_mdl(mdl, df = fit_df, rsp_var, rsp_var_out,
## model_id_method, : Expecting 1 metric: Accuracy; recd: Accuracy, Kappa;
## retaining Accuracy only
## model_id model_method
## 1 Final.rf rf
## feats
## 1 T.phase, A.toxic, A.mgm2, A.surviv, T.versus, A.respons, A.median, A.rate, T.num.chars, A.cyclophosphamid, T.advanc, A.num.chars, T.breast, A.num.words, A.num.words.unq, A.predict, T.num.words.unq, T.num.words, T.cyclophosphamid, A.use, A.everi, T.chemotherapi, A.risk, A.cell, A.tamoxifen, T.studi, A.month, A.reduc, A.chemotherapi, A.firstlin, A.placebo, A.patient, A.prevent, A.control, A.trial, T.respons, A.day, A.arm, A.durat, A.receiv, A.dose, A.breast, A.treatment, A.increas, A.data, A.bone, A.either, T.tamoxifen, A.fluorouracil, A.year, A.toler, A.alon, A.associ, A.chang, A.methotrex, A.radiotherapi, A.metastas, A.cycl, A.combin, A.efficaci, A.postop, A.compar, A.overal, A.cmf, A.regimen, A.advanc, A.week, A.seen, A.women, A.improv, A.cours, A.random, A.baselin, T.cancer, T.metastat, A.term, T.adjuv, A.given, A.among, A.two, A.cancer, A.group, A.diseasefre, A.high, A.therapi, A.differ, A.progress, A.clinic, A.oral, A.daili, A.total, T.compar, T.combin, T.treatment, A.analysi, A.identifi, A.signific, T.patient, A.effect, A.postmenopaus, A.relaps, A.studi, T.women, A.superior, A.measur, A.prognost, A.factor, A.time, A.surgeri, A.evid, A.intraven, A.stage, T.clinic, T.result, T.trial, A.level, A.pretreat, A.X100, A.assess, A.recurr, A.life, A.system, A.event, A.dfs, A.obtain, T.effect, A.without, T.group, A.randomis, T.randomis, A.nodeposit, A.previous, A.adjuv, A.followup, A.statist, A.primari, A.first, A.treat, A.assign, A.develop, A.benefit, A.activ, A.show, A.carcinoma, A.analyz, A.result, A.doxorubicin, A.respond, A.agent, A.plus, A.administ, A.versus, A.follow, A.achiev, A.failur, A.metastat, A.standard, T.plus, A.decreas, A.iii, A.consid, A.administr, A.complet, A.tumor, A.express, A.hundr, A.suggest, A.estrogen, A.site, A.respect, A.endpoint, A.three, A.paclitaxel, A.observ, A.wherea, A.sequenti, A.oper, A.schedul, A.side, A.evalu, A.reduct, A.analys, A.epirubicin, A.earli, T.earli, A.X5fluorouracil, A.axillari, A.anthracyclin, A.conduct, A.advers, A.phase, A.receptor, A.status, A.lymph, A.initi, A.receptorposit, A.provid, A.outcom, T.postmenopaus, A.perform, A.multivari, A.safeti, T.therapi, A.qualiti, A.hormon, A.rang, A.caus, A.confid, A.serum, A.appear, A.docetaxel, A.four, A.diseas, A.model, A.conclus, A.prolong, A.independ, A.one, A.continu, A.may, T.random, A.object, A.found, A.period, A.report, A.enter, A.within, A.low, A.progressionfre, A.similar, A.although, A.indic, A.X001, A.occur, A.longer, A.larg, A.addit, A.endocrin, A.correl, A.inform, A.partial, A.involv, A.function., A.drug, A.examin, A.incid, A.method, A.select, A.determin, A.age, A.X500, A.import, A.nausea, A.posit, A.also, A.premenopaus, A.avail, A.progesteron, A.less, A.mastectomi, A.whether, A.better, A.end, A.tissu, A.neoadjuv, A.interv, A.investig, A.can, A.lower, A.node, A.prospect, A.X005, A.purpos, A.second, A.trend, A.aim, A.distant, A.ratio, A.stabl, A.mean, A.popul, A.infus, A.shown, A.includ, A.possibl, A.common, A.hazard, A.present, A.number, T.doxorubicin, A.proport, A.consist, A.well, A.remain, A.mbc, A.doubleblind, A.accord, A.local, A.higher, A.start, A.relat, A.tumour, A.design, A.secondari, A.confirm, A.demonstr, A.estim, A.negat, A.main, A.sampl, A.six, A.base, A.sever, A.per, A.find, A.growth, T.iii, A.enrol, A.elig, A.neutropenia, A.valu, A.howev, A.vomit, A.least, A.X0001, A.greater, A.death, T.docetaxel, A.due, A.grade, A.case, A.marker, A.major, A.experienc, A.frequent, A.need, A.patholog, A.singl, A.comparison, A.support, A.prior, A.eight, A.particip, A.histolog, A.potenti, A.subgroup, A.inhibitor, A.detect, A.test, A.background, A.point, A.her2, A.size, A.seven, A.five, A.requir, A.new, A.type, A.andor, A.set, A.regard, A.score, A.general, A.regress, A.limit, A.defin, A.aromatas, A.multicent, A.characterist, A.human, A.profil, A.hematolog
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 66.931 32.821
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 1 0.6 1 0.8417819
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit
## 1 0.9971708 1 0.6770009glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.data.training.all",
chunk_step_major=glb_script_df[nrow(glb_script_df), "chunk_step_major"],
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed11 fit.data.training.all 6 0 440.471
## elapsed12 fit.data.training.all 6 1 511.814glb_rsp_var_out <- paste0(glb_rsp_var_out, tail(names(glb_models_lst), 1))
# Used again in predict.data.new chunk
glb_get_predictions <- function(df) {
if (glb_is_regression) {
df[, glb_rsp_var_out] <- predict(glb_fin_mdl, newdata=df, type="raw")
print(myplot_scatter(df, glb_rsp_var, glb_rsp_var_out,
smooth=TRUE))
df[, paste0(glb_rsp_var_out, ".err")] <-
abs(df[, glb_rsp_var_out] - df[, glb_rsp_var])
print(head(orderBy(reformulate(c("-", paste0(glb_rsp_var_out, ".err"))),
df)))
}
if (glb_is_classification && glb_is_binomial) {
# incorporate glb_clf_proba_threshold
# shd it only be for glb_fin_mdl or for earlier models ?
# for glb_trnent_df it shd opt; vs. assume for glb_newent_df
finmdl_prob_fit <- glb_models_df[glb_models_df$model_id == glb_fin_mdl_id,
"opt.prob.threshold.fit"]
selmdl_prob_fit <- glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.fit"]
if (finmdl_prob_fit != selmdl_prob_fit)
warning("opt.prob.threshold.fit differs for fin_mdl: ", finmdl_prob_fit,
" vs. sel_mdl: ", selmdl_prob_fit)
prob_threshold <- glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"]
df[, paste0(glb_rsp_var_out, ".prob")] <-
predict(glb_fin_mdl, newdata=df, type="prob")[, 2]
df[, glb_rsp_var_out] <-
factor(levels(df[, glb_rsp_var])[
(df[, paste0(glb_rsp_var_out, ".prob")] >=
prob_threshold) * 1 + 1], levels(df[, glb_rsp_var]))
# prediction stats already reported by myfit_mdl ???
}
if (glb_is_classification && !glb_is_binomial) {
df[, glb_rsp_var_out] <- predict(glb_fin_mdl, newdata=df, type="raw")
}
return(df)
}
glb_trnent_df <- glb_get_predictions(df=glb_trnent_df)
print(glb_feats_df <- mymerge_feats_importance(feats_df=glb_feats_df, sel_mdl=glb_fin_mdl,
entity_df=glb_trnent_df))## id cor.y exclude.as.feat cor.y.abs
## 362 T.phase 4.407918e-01 0 4.407918e-01
## 314 A.toxic 3.546830e-01 0 3.546830e-01
## 181 A.mgm2 3.476197e-01 0 3.476197e-01
## 303 A.surviv 1.844165e-01 0 1.844165e-01
## 266 A.respons 2.450234e-01 0 2.450234e-01
## 176 A.median 2.890542e-01 0 2.890542e-01
## 374 T.versus 2.497154e-01 0 2.497154e-01
## 358 T.num.chars 1.122127e-01 0 1.122127e-01
## 65 A.cyclophosphamid 1.958128e-01 0 1.958128e-01
## 248 A.rate 2.658285e-01 0 2.658285e-01
## 194 A.num.chars -7.349760e-03 0 7.349760e-03
## 342 T.advanc 2.469661e-01 0 2.469661e-01
## 343 T.breast 1.989942e-01 0 1.989942e-01
## 195 A.num.words -1.799787e-02 0 1.799787e-02
## 225 A.predict -1.756009e-01 0 1.756009e-01
## 196 A.num.words.unq 1.784364e-02 0 1.784364e-02
## 359 T.num.words 1.040679e-01 0 1.040679e-01
## 105 A.everi 2.932475e-01 0 2.932475e-01
## 268 A.risk -1.845209e-01 0 1.845209e-01
## 360 T.num.words.unq 8.542797e-02 0 8.542797e-02
## 345 T.chemotherapi -9.909758e-03 0 9.909758e-03
## 349 T.cyclophosphamid 1.932487e-01 0 1.932487e-01
## 323 A.use -2.345315e-01 0 2.345315e-01
## 254 A.reduc -1.472907e-01 0 1.472907e-01
## 305 A.tamoxifen 7.004532e-03 0 7.004532e-03
## 43 A.cell -1.478798e-01 0 1.478798e-01
## 369 T.studi 2.657631e-01 0 2.657631e-01
## 114 A.firstlin 2.039271e-01 0 2.039271e-01
## 216 A.placebo -1.229875e-01 0 1.229875e-01
## 211 A.patient 1.420881e-01 0 1.420881e-01
## 312 A.toler 2.093203e-01 0 2.093203e-01
## 183 A.month 1.887787e-01 0 1.887787e-01
## 316 A.treatment 6.578098e-02 0 6.578098e-02
## 367 T.respons -1.496979e-01 0 1.496979e-01
## 229 A.prevent -1.556778e-01 0 1.556778e-01
## 46 A.chemotherapi -6.848646e-02 0 6.848646e-02
## 318 A.trial -1.258125e-01 0 1.258125e-01
## 93 A.either 1.309481e-01 0 1.309481e-01
## 88 A.durat 1.644884e-01 0 1.644884e-01
## 250 A.receiv 6.090943e-02 0 6.090943e-02
## 83 A.dose 1.284812e-01 0 1.284812e-01
## 146 A.increas -1.192067e-01 0 1.192067e-01
## 36 A.bone -1.552270e-01 0 1.552270e-01
## 61 A.control -1.360290e-01 0 1.360290e-01
## 340 A.year 9.566746e-03 0 9.566746e-03
## 116 A.fluorouracil 1.156599e-01 0 1.156599e-01
## 68 A.day 1.770836e-01 0 1.770836e-01
## 14 A.alon 3.907515e-02 0 3.907515e-02
## 37 A.breast -5.233490e-02 0 5.233490e-02
## 24 A.arm 1.729602e-01 0 1.729602e-01
## 28 A.associ -1.738833e-01 0 1.738833e-01
## 80 A.diseasefre 9.577232e-02 0 9.577232e-02
## 91 A.efficaci 1.446412e-01 0 1.446412e-01
## 206 A.overal 1.927655e-01 0 1.927655e-01
## 275 A.seen 1.203577e-01 0 1.203577e-01
## 370 T.tamoxifen 5.564564e-02 0 5.564564e-02
## 51 A.compar 5.923358e-02 0 5.923358e-02
## 245 A.random 3.663241e-02 0 3.663241e-02
## 48 A.cmf 1.023838e-01 0 1.023838e-01
## 177 A.metastas -8.500380e-02 0 8.500380e-02
## 348 T.compar 1.658208e-01 0 1.658208e-01
## 67 A.data -1.605586e-01 0 1.605586e-01
## 257 A.regimen 2.028112e-01 0 2.028112e-01
## 306 A.term 7.818056e-02 0 7.818056e-02
## 244 A.radiotherapi -7.542001e-02 0 7.542001e-02
## 64 A.cycl 1.511902e-01 0 1.511902e-01
## 133 A.high -1.285034e-01 0 1.285034e-01
## 49 A.combin 2.088035e-01 0 2.088035e-01
## 124 A.given 1.577815e-01 0 1.577815e-01
## 333 A.women -6.706568e-02 0 6.706568e-02
## 140 A.identifi -1.524843e-01 0 1.524843e-01
## 180 A.methotrex 1.508929e-01 0 1.508929e-01
## 285 A.signific -7.475443e-02 0 7.475443e-02
## 33 A.baselin -1.517685e-01 0 1.517685e-01
## 143 A.improv 7.017871e-03 0 7.017871e-03
## 313 A.total 5.916846e-02 0 5.916846e-02
## 66 A.daili 6.879414e-02 0 6.879414e-02
## 44 A.chang -1.494941e-01 0 1.494941e-01
## 127 A.group -1.337887e-02 0 1.337887e-02
## 47 A.clinic -7.577756e-02 0 7.577756e-02
## 321 A.two 1.118305e-01 0 1.118305e-01
## 63 A.cours 5.392045e-02 0 5.392045e-02
## 341 T.adjuv -2.813822e-02 0 2.813822e-02
## 372 T.treatment 5.769991e-02 0 5.769991e-02
## 357 T.metastat 3.049464e-01 0 3.049464e-01
## 308 A.therapi -2.549221e-02 0 2.549221e-02
## 222 A.postmenopaus 2.765137e-02 0 2.765137e-02
## 223 A.postop -2.083001e-02 0 2.083001e-02
## 17 A.among 3.632848e-04 0 3.632848e-04
## 161 A.life -7.741262e-02 0 7.741262e-02
## 204 A.oral 9.728974e-02 0 9.728974e-02
## 327 A.week 1.918602e-01 0 1.918602e-01
## 39 A.cancer -5.581714e-02 0 5.581714e-02
## 236 A.progress 2.284989e-01 0 2.284989e-01
## 78 A.differ 1.797677e-02 0 1.797677e-02
## 337 A.X100 1.513446e-01 0 1.513446e-01
## 118 A.followup 2.973279e-02 0 2.973279e-02
## 175 A.measur -1.483961e-01 0 1.483961e-01
## 310 A.time 6.178757e-02 0 6.178757e-02
## 304 A.system -8.347057e-02 0 8.347057e-02
## 347 T.combin 1.835246e-01 0 1.835246e-01
## 375 T.women -7.520353e-02 0 7.520353e-02
## 9 A.advanc 2.281995e-01 0 2.281995e-01
## 292 A.stage 3.975581e-02 0 3.975581e-02
## 253 A.recurr -6.713675e-02 0 6.713675e-02
## 160 A.level -1.206844e-01 0 1.206844e-01
## 267 A.result 1.442596e-02 0 1.442596e-02
## 90 A.effect -7.906464e-02 0 7.906464e-02
## 366 T.randomis 9.370110e-02 0 9.370110e-02
## 361 T.patient 7.667604e-03 0 7.667604e-03
## 344 T.cancer 1.680898e-01 0 1.680898e-01
## 297 A.studi -2.594717e-02 0 2.594717e-02
## 259 A.relaps -7.929870e-06 0 7.929870e-06
## 300 A.superior 1.144624e-01 0 1.144624e-01
## 26 A.assess -1.157562e-01 0 1.157562e-01
## 346 T.clinic -5.148394e-02 0 5.148394e-02
## 217 A.plus 1.174459e-01 0 1.174459e-01
## 354 T.group 1.488481e-01 0 1.488481e-01
## 104 A.event 8.489612e-02 0 8.489612e-02
## 110 A.factor -1.049804e-01 0 1.049804e-01
## 235 A.prognost -1.363777e-01 0 1.363777e-01
## 6 A.adjuv -5.608234e-02 0 5.608234e-02
## 373 T.trial 1.195178e-01 0 1.195178e-01
## 228 A.pretreat 4.965851e-02 0 4.965851e-02
## 27 A.assign 7.540280e-02 0 7.540280e-02
## 295 A.statist 1.141908e-03 0 1.141908e-03
## 193 A.nodeposit 7.498902e-02 0 7.498902e-02
## 70 A.decreas -1.536759e-01 0 1.536759e-01
## 315 A.treat 1.103432e-02 0 1.103432e-02
## 302 A.surgeri -2.256826e-03 0 2.256826e-03
## 299 A.suggest -5.757780e-02 0 5.757780e-02
## 178 A.metastat 2.417899e-01 0 2.417899e-01
## 12 A.agent 3.257901e-02 0 3.257901e-02
## 19 A.analysi -8.745562e-02 0 8.745562e-02
## 106 A.evid 5.914321e-03 0 5.914321e-03
## 7 A.administ 7.791871e-02 0 7.791871e-02
## 85 A.doxorubicin 1.217472e-01 0 1.217472e-01
## 282 A.show -9.265780e-02 0 9.265780e-02
## 200 A.obtain -5.756790e-02 0 5.756790e-02
## 111 A.failur 7.764819e-02 0 7.764819e-02
## 113 A.first -7.206416e-02 0 7.206416e-02
## 332 A.without -1.584596e-03 0 1.584596e-03
## 368 T.result 5.427221e-02 0 5.427221e-02
## 89 A.earli -3.276589e-02 0 3.276589e-02
## 231 A.primari 6.020830e-02 0 6.020830e-02
## 102 A.estrogen -5.357349e-02 0 5.357349e-02
## 264 A.respect 1.158080e-01 0 1.158080e-01
## 353 T.effect -1.798643e-01 0 1.798643e-01
## 154 A.intraven 5.564821e-02 0 5.564821e-02
## 40 A.carcinoma -3.493790e-02 0 3.493790e-02
## 277 A.sequenti 9.916268e-02 0 9.916268e-02
## 117 A.follow 6.707377e-02 0 6.707377e-02
## 246 A.randomis 1.425652e-02 0 1.425652e-02
## 185 A.multivari -1.231479e-01 0 1.231479e-01
## 271 A.schedul 1.358289e-01 0 1.358289e-01
## 76 A.develop -3.000773e-02 0 3.000773e-02
## 325 A.versus 5.026899e-02 0 5.026899e-02
## 77 A.dfs 6.076035e-02 0 6.076035e-02
## 139 A.hundr 1.362355e-01 0 1.362355e-01
## 4 A.activ 1.271691e-01 0 1.271691e-01
## 269 A.safeti 1.217685e-01 0 1.217685e-01
## 100 A.epirubicin 1.576660e-01 0 1.576660e-01
## 8 A.administr 2.040770e-02 0 2.040770e-02
## 34 A.benefit -9.524764e-03 0 9.524764e-03
## 141 A.iii 3.062659e-03 0 3.062659e-03
## 97 A.endpoint 7.093389e-02 0 7.093389e-02
## 213 A.perform -4.866672e-02 0 4.866672e-02
## 58 A.consid -1.417986e-02 0 1.417986e-02
## 56 A.confid 7.049294e-02 0 7.049294e-02
## 309 A.three 1.031348e-01 0 1.031348e-01
## 199 A.observ 5.273205e-02 0 5.273205e-02
## 293 A.standard 2.641969e-02 0 2.641969e-02
## 20 A.analyz -1.060879e-01 0 1.060879e-01
## 53 A.complet 9.717265e-02 0 9.717265e-02
## 215 A.phase 1.481848e-01 0 1.481848e-01
## 364 T.postmenopaus 4.801689e-02 0 4.801689e-02
## 3 A.achiev 9.514020e-02 0 9.514020e-02
## 103 A.evalu 5.098452e-02 0 5.098452e-02
## 109 A.express -1.321424e-01 0 1.321424e-01
## 288 A.site 7.553580e-03 0 7.553580e-03
## 339 A.X5fluorouracil 1.153122e-01 0 1.153122e-01
## 203 A.oper 3.993046e-02 0 3.993046e-02
## 265 A.respond 7.372052e-02 0 7.372052e-02
## 11 A.age -3.188645e-02 0 3.188645e-02
## 30 A.axillari -1.556611e-02 0 1.556611e-02
## 153 A.interv 5.928858e-02 0 5.928858e-02
## 243 A.qualiti -1.007081e-01 0 1.007081e-01
## 119 A.found -1.328394e-01 0 1.328394e-01
## 230 A.previous 1.535045e-01 0 1.535045e-01
## 241 A.provid -9.854909e-02 0 9.854909e-02
## 207 A.paclitaxel 1.044681e-01 0 1.044681e-01
## 319 A.tumor -9.949416e-02 0 9.949416e-02
## 247 A.rang 1.050787e-01 0 1.050787e-01
## 145 A.includ 4.584644e-02 0 4.584644e-02
## 42 A.caus -1.915695e-02 0 1.915695e-02
## 22 A.anthracyclin 8.474734e-02 0 8.474734e-02
## 201 A.occur 1.363943e-01 0 1.363943e-01
## 329 A.wherea -8.264291e-02 0 8.264291e-02
## 284 A.side 2.689790e-02 0 2.689790e-02
## 38 A.can -1.009582e-01 0 1.009582e-01
## 79 A.diseas 1.454492e-01 0 1.454492e-01
## 120 A.four 1.512749e-01 0 1.512749e-01
## 136 A.hormon -5.048861e-02 0 5.048861e-02
## 18 A.analys -8.879582e-02 0 8.879582e-02
## 148 A.indic -8.330481e-02 0 8.330481e-02
## 255 A.reduct -5.075505e-02 0 5.075505e-02
## 296 A.status -4.120076e-02 0 4.120076e-02
## 262 A.report -8.926696e-02 0 8.926696e-02
## 252 A.receptorposit 2.492618e-02 0 2.492618e-02
## 202 A.one 8.221967e-02 0 8.221967e-02
## 286 A.similar 9.371701e-02 0 9.371701e-02
## 15 A.also -7.612550e-02 0 7.612550e-02
## 147 A.independ -1.181013e-01 0 1.181013e-01
## 208 A.partial 2.185387e-01 0 2.185387e-01
## 352 T.earli -7.039691e-02 0 7.039691e-02
## 149 A.inform -1.140931e-01 0 1.140931e-01
## 251 A.receptor -2.389969e-02 0 2.389969e-02
## 152 A.initi -1.332423e-02 0 1.332423e-02
## 198 A.object 1.065300e-01 0 1.065300e-01
## 363 T.plus 1.614591e-01 0 1.614591e-01
## 338 A.X500 1.433281e-01 0 1.433281e-01
## 205 A.outcom -8.083204e-02 0 8.083204e-02
## 86 A.drug -2.504858e-03 0 2.504858e-03
## 351 T.doxorubicin 1.324163e-01 0 1.324163e-01
## 55 A.conduct 2.467401e-02 0 2.467401e-02
## 62 A.correl -1.185756e-01 0 1.185756e-01
## 371 T.therapi -3.064253e-03 0 3.064253e-03
## 283 A.shown -4.986612e-02 0 4.986612e-02
## 165 A.low -7.478252e-02 0 7.478252e-02
## 226 A.premenopaus -6.600118e-03 0 6.600118e-03
## 5 A.addit 3.510086e-02 0 3.510086e-02
## 278 A.serum -1.037758e-01 0 1.037758e-01
## 238 A.prolong 6.968693e-02 0 6.968693e-02
## 144 A.incid -4.227791e-02 0 4.227791e-02
## 167 A.lymph 3.005185e-02 0 3.005185e-02
## 82 A.docetaxel 1.721916e-01 0 1.721916e-01
## 16 A.although 3.604544e-02 0 3.604544e-02
## 23 A.appear 1.314392e-02 0 1.314392e-02
## 237 A.progressionfre 1.270990e-01 0 1.270990e-01
## 227 A.present -4.695298e-02 0 4.695298e-02
## 29 A.avail -8.834244e-02 0 8.834244e-02
## 365 T.random 1.028374e-01 0 1.028374e-01
## 10 A.advers 5.167782e-02 0 5.167782e-02
## 182 A.model -1.234019e-01 0 1.234019e-01
## 122 A.function. -8.739253e-02 0 8.739253e-02
## 96 A.endocrin -1.936997e-02 0 1.936997e-02
## 171 A.mastectomi -1.359074e-02 0 1.359074e-02
## 172 A.may -8.289575e-02 0 8.289575e-02
## 240 A.prospect -3.102924e-02 0 3.102924e-02
## 214 A.period -5.074228e-02 0 5.074228e-02
## 242 A.purpos -6.501960e-03 0 6.501960e-03
## 291 A.stabl 7.053656e-02 0 7.053656e-02
## 75 A.determin -4.239248e-02 0 4.239248e-02
## 35 A.better 4.133124e-02 0 4.133124e-02
## 60 A.continu 6.889854e-02 0 6.889854e-02
## 334 A.X0001 -6.383307e-02 0 6.383307e-02
## 142 A.import -6.500431e-02 0 6.500431e-02
## 159 A.less 3.732604e-02 0 3.732604e-02
## 335 A.X001 -3.311267e-02 0 3.311267e-02
## 219 A.popul -1.296710e-02 0 1.296710e-02
## 336 A.X005 -8.880757e-02 0 8.880757e-02
## 273 A.second -5.198940e-02 0 5.198940e-02
## 179 A.method -3.582004e-02 0 3.582004e-02
## 13 A.aim -4.785872e-02 0 4.785872e-02
## 107 A.examin -1.361521e-01 0 1.361521e-01
## 320 A.tumour -1.029714e-01 0 1.029714e-01
## 163 A.local -4.366301e-02 0 4.366301e-02
## 331 A.within -5.934658e-02 0 5.934658e-02
## 164 A.longer 8.421948e-02 0 8.421948e-02
## 326 A.vomit 1.161292e-01 0 1.161292e-01
## 239 A.proport -1.076947e-01 0 1.076947e-01
## 328 A.well 8.870282e-02 0 8.870282e-02
## 330 A.whether -1.291745e-02 0 1.291745e-02
## 32 A.base -3.357553e-02 0 3.357553e-02
## 41 A.case -5.636252e-02 0 5.636252e-02
## 99 A.enter 1.130515e-01 0 1.130515e-01
## 84 A.doubleblind -1.009869e-01 0 1.009869e-01
## 197 A.number -2.963870e-02 0 2.963870e-02
## 317 A.trend -2.337564e-02 0 2.337564e-02
## 150 A.infus -2.691231e-03 0 2.691231e-03
## 155 A.investig -6.814008e-02 0 6.814008e-02
## 95 A.end 6.542423e-02 0 6.542423e-02
## 50 A.common 7.737623e-02 0 7.737623e-02
## 298 A.subgroup -2.338162e-02 0 2.338162e-02
## 69 A.death 8.448898e-02 0 8.448898e-02
## 156 A.involv 2.700336e-02 0 2.700336e-02
## 186 A.nausea 4.537254e-02 0 4.537254e-02
## 72 A.demonstr 1.364004e-03 0 1.364004e-03
## 81 A.distant -2.186214e-02 0 2.186214e-02
## 166 A.lower -6.958213e-02 0 6.958213e-02
## 132 A.her2 -4.085805e-02 0 4.085805e-02
## 125 A.grade 1.432705e-01 0 1.432705e-01
## 157 A.larg -4.945536e-02 0 4.945536e-02
## 276 A.select -7.095647e-02 0 7.095647e-02
## 274 A.secondari 8.944036e-02 0 8.944036e-02
## 174 A.mean -7.485297e-02 0 7.485297e-02
## 54 A.conclus 2.541412e-03 0 2.541412e-03
## 224 A.potenti -6.439343e-02 0 6.439343e-02
## 220 A.posit -4.317584e-02 0 4.317584e-02
## 130 A.hazard 1.634076e-02 0 1.634076e-02
## 94 A.elig 1.201664e-01 0 1.201664e-01
## 31 A.background -9.175455e-03 0 9.175455e-03
## 221 A.possibl -4.794828e-02 0 4.794828e-02
## 137 A.howev -2.671381e-02 0 2.671381e-02
## 294 A.start -4.140127e-02 0 4.140127e-02
## 73 A.design -6.374264e-02 0 6.374264e-02
## 311 A.tissu -5.595253e-02 0 5.595253e-02
## 192 A.node -1.504773e-02 0 1.504773e-02
## 134 A.higher -3.196262e-03 0 3.196262e-03
## 59 A.consist 1.855504e-02 0 1.855504e-02
## 190 A.neutropenia 2.262877e-01 0 2.262877e-01
## 115 A.five 4.460546e-02 0 4.460546e-02
## 2 A.accord 9.173456e-03 0 9.173456e-03
## 324 A.valu -1.111873e-01 0 1.111873e-01
## 112 A.find -5.637269e-02 0 5.637269e-02
## 151 A.inhibitor -1.239898e-02 0 1.239898e-02
## 249 A.ratio 6.872134e-03 0 6.872134e-03
## 108 A.experienc 6.719543e-02 0 6.719543e-02
## 168 A.main -4.489330e-02 0 4.489330e-02
## 52 A.comparison -3.040785e-02 0 3.040785e-02
## 287 A.singl 1.695022e-02 0 1.695022e-02
## 126 A.greater -2.407688e-02 0 2.407688e-02
## 187 A.need -1.172494e-01 0 1.172494e-01
## 173 A.mbc 1.540985e-01 0 1.540985e-01
## 356 T.iii 1.663443e-01 0 1.663443e-01
## 234 A.progesteron -4.318635e-02 0 4.318635e-02
## 301 A.support -2.793158e-02 0 2.793158e-02
## 158 A.least -1.387475e-02 0 1.387475e-02
## 98 A.enrol 7.928104e-02 0 7.928104e-02
## 189 A.neoadjuv 2.210203e-02 0 2.210203e-02
## 261 A.remain -6.292028e-02 0 6.292028e-02
## 218 A.point 4.946137e-02 0 4.946137e-02
## 289 A.six 9.362652e-02 0 9.362652e-02
## 128 A.growth -9.887605e-03 0 9.887605e-03
## 57 A.confirm -1.221603e-02 0 1.221603e-02
## 260 A.relat -7.729160e-02 0 7.729160e-02
## 188 A.negat -8.863047e-02 0 8.863047e-02
## 170 A.marker -1.391338e-01 0 1.391338e-01
## 21 A.andor 3.086932e-02 0 3.086932e-02
## 232 A.prior 1.026878e-01 0 1.026878e-01
## 307 A.test -8.422639e-02 0 8.422639e-02
## 92 A.eight 8.362672e-02 0 8.362672e-02
## 281 A.sever 1.516401e-02 0 1.516401e-02
## 263 A.requir -2.281593e-02 0 2.281593e-02
## 191 A.new -2.033234e-02 0 2.033234e-02
## 25 A.aromatas -4.322202e-02 0 4.322202e-02
## 258 A.regress -3.139311e-02 0 3.139311e-02
## 210 A.patholog -3.751957e-02 0 3.751957e-02
## 350 T.docetaxel 1.947844e-01 0 1.947844e-01
## 169 A.major 2.209530e-02 0 2.209530e-02
## 270 A.sampl -1.119119e-01 0 1.119119e-01
## 87 A.due 3.164831e-02 0 3.164831e-02
## 101 A.estim -1.960142e-02 0 1.960142e-02
## 135 A.histolog -6.904065e-02 0 6.904065e-02
## 290 A.size -7.746176e-02 0 7.746176e-02
## 279 A.set -3.997140e-02 0 3.997140e-02
## 212 A.per -3.173574e-02 0 3.173574e-02
## 121 A.frequent 7.708184e-02 0 7.708184e-02
## 209 A.particip -1.330350e-01 0 1.330350e-01
## 162 A.limit 1.884501e-02 0 1.884501e-02
## 138 A.human -9.121557e-04 0 9.121557e-04
## 233 A.profil 1.092909e-02 0 1.092909e-02
## 74 A.detect -6.580512e-02 0 6.580512e-02
## 256 A.regard -2.212204e-02 0 2.212204e-02
## 123 A.general -3.234203e-02 0 3.234203e-02
## 280 A.seven 4.541994e-02 0 4.541994e-02
## 272 A.score -1.320092e-01 0 1.320092e-01
## 45 A.characterist -1.258143e-02 0 1.258143e-02
## 184 A.multicent 4.053909e-02 0 4.053909e-02
## 71 A.defin -2.332432e-02 0 2.332432e-02
## 131 A.hematolog 1.187144e-01 0 1.187144e-01
## 322 A.type -8.602559e-02 0 8.602559e-02
## 1 .rnorm 4.466651e-02 0 4.466651e-02
## 129 A.has.http -3.472002e-02 0 3.472002e-02
## 355 T.has.http NA 0 NA
## 376 trial 1.000000e+00 1 1.000000e+00
## cor.high.X is.ConditionalX.y is.cor.y.abs.low importance
## 362 <NA> TRUE FALSE 1.000000e+02
## 314 <NA> TRUE FALSE 7.242079e+01
## 181 <NA> TRUE FALSE 4.648562e+01
## 303 <NA> TRUE FALSE 2.686299e+01
## 266 <NA> TRUE FALSE 1.923775e+01
## 176 <NA> TRUE FALSE 1.839560e+01
## 374 <NA> TRUE FALSE 1.604633e+01
## 358 T.num.words TRUE FALSE 1.207477e+01
## 65 <NA> TRUE FALSE 1.042900e+01
## 248 <NA> TRUE FALSE 9.754765e+00
## 194 <NA> TRUE TRUE 9.195501e+00
## 342 <NA> TRUE FALSE 8.986921e+00
## 343 T.cancer TRUE FALSE 8.593111e+00
## 195 <NA> TRUE TRUE 8.081324e+00
## 225 <NA> TRUE FALSE 7.933750e+00
## 196 <NA> TRUE TRUE 7.926826e+00
## 359 T.num.words.unq TRUE FALSE 6.515790e+00
## 105 <NA> TRUE FALSE 6.217728e+00
## 268 <NA> TRUE FALSE 5.772376e+00
## 360 <NA> TRUE FALSE 5.772069e+00
## 345 <NA> TRUE TRUE 5.383444e+00
## 349 <NA> TRUE FALSE 5.156588e+00
## 323 <NA> TRUE FALSE 4.981764e+00
## 254 <NA> TRUE FALSE 4.872754e+00
## 305 <NA> TRUE TRUE 4.768437e+00
## 43 <NA> TRUE FALSE 4.087102e+00
## 369 <NA> TRUE FALSE 4.015622e+00
## 114 <NA> TRUE FALSE 3.876175e+00
## 216 <NA> TRUE FALSE 3.874297e+00
## 211 <NA> TRUE FALSE 3.867235e+00
## 312 <NA> TRUE FALSE 3.818460e+00
## 183 <NA> TRUE FALSE 3.773358e+00
## 316 <NA> TRUE FALSE 3.590879e+00
## 367 <NA> TRUE FALSE 3.583397e+00
## 229 <NA> TRUE FALSE 3.543041e+00
## 46 <NA> TRUE FALSE 3.433307e+00
## 318 <NA> TRUE FALSE 3.257179e+00
## 93 <NA> TRUE FALSE 3.153436e+00
## 88 <NA> TRUE FALSE 3.101665e+00
## 250 <NA> TRUE FALSE 3.093108e+00
## 83 <NA> TRUE FALSE 3.044180e+00
## 146 <NA> TRUE FALSE 2.983883e+00
## 36 <NA> TRUE FALSE 2.904102e+00
## 61 <NA> TRUE FALSE 2.889240e+00
## 340 <NA> TRUE TRUE 2.824132e+00
## 116 <NA> TRUE FALSE 2.804778e+00
## 68 <NA> TRUE FALSE 2.790640e+00
## 14 <NA> TRUE TRUE 2.698542e+00
## 37 <NA> TRUE FALSE 2.664569e+00
## 24 <NA> TRUE FALSE 2.599298e+00
## 28 <NA> TRUE FALSE 2.583765e+00
## 80 <NA> TRUE FALSE 2.577309e+00
## 91 <NA> TRUE FALSE 2.549580e+00
## 206 <NA> TRUE FALSE 2.539388e+00
## 275 <NA> TRUE FALSE 2.529300e+00
## 370 <NA> TRUE FALSE 2.525166e+00
## 51 <NA> TRUE FALSE 2.436673e+00
## 245 <NA> TRUE TRUE 2.434623e+00
## 48 <NA> TRUE FALSE 2.421882e+00
## 177 <NA> TRUE FALSE 2.349916e+00
## 348 <NA> TRUE FALSE 2.308743e+00
## 67 <NA> TRUE FALSE 2.278653e+00
## 257 <NA> TRUE FALSE 2.266469e+00
## 306 <NA> TRUE FALSE 2.263556e+00
## 244 <NA> TRUE FALSE 2.244227e+00
## 64 <NA> TRUE FALSE 2.176275e+00
## 133 <NA> TRUE FALSE 2.176216e+00
## 49 <NA> TRUE FALSE 2.174240e+00
## 124 <NA> TRUE FALSE 2.171904e+00
## 333 <NA> TRUE FALSE 2.148318e+00
## 140 <NA> TRUE FALSE 2.070562e+00
## 180 <NA> TRUE FALSE 2.004642e+00
## 285 <NA> TRUE FALSE 2.004394e+00
## 33 <NA> TRUE FALSE 1.993126e+00
## 143 <NA> TRUE TRUE 1.987598e+00
## 313 <NA> TRUE FALSE 1.962843e+00
## 66 <NA> TRUE FALSE 1.958864e+00
## 44 <NA> TRUE FALSE 1.943340e+00
## 127 <NA> TRUE TRUE 1.901921e+00
## 47 <NA> TRUE FALSE 1.861954e+00
## 321 <NA> TRUE FALSE 1.841165e+00
## 63 <NA> TRUE FALSE 1.834601e+00
## 341 <NA> TRUE TRUE 1.819596e+00
## 372 <NA> TRUE FALSE 1.798402e+00
## 357 <NA> TRUE FALSE 1.776861e+00
## 308 <NA> TRUE TRUE 1.768894e+00
## 222 <NA> TRUE TRUE 1.759164e+00
## 223 <NA> TRUE TRUE 1.707956e+00
## 17 <NA> TRUE TRUE 1.701172e+00
## 161 <NA> TRUE FALSE 1.687285e+00
## 204 <NA> TRUE FALSE 1.678807e+00
## 327 <NA> TRUE FALSE 1.669876e+00
## 39 A.breast TRUE FALSE 1.660957e+00
## 236 <NA> TRUE FALSE 1.647024e+00
## 78 <NA> TRUE TRUE 1.644295e+00
## 337 <NA> TRUE FALSE 1.633285e+00
## 118 <NA> TRUE TRUE 1.623447e+00
## 175 <NA> TRUE FALSE 1.603485e+00
## 310 <NA> TRUE FALSE 1.576943e+00
## 304 <NA> TRUE FALSE 1.548913e+00
## 347 <NA> TRUE FALSE 1.536732e+00
## 375 <NA> TRUE FALSE 1.524973e+00
## 9 <NA> TRUE FALSE 1.517969e+00
## 292 <NA> TRUE TRUE 1.514853e+00
## 253 <NA> TRUE FALSE 1.470009e+00
## 160 <NA> TRUE FALSE 1.450451e+00
## 267 <NA> TRUE TRUE 1.448685e+00
## 90 <NA> TRUE FALSE 1.441340e+00
## 366 <NA> TRUE FALSE 1.438026e+00
## 361 <NA> TRUE TRUE 1.424822e+00
## 344 <NA> TRUE FALSE 1.405536e+00
## 297 <NA> TRUE TRUE 1.399510e+00
## 259 <NA> TRUE TRUE 1.394143e+00
## 300 <NA> TRUE FALSE 1.391479e+00
## 26 <NA> TRUE FALSE 1.388645e+00
## 346 <NA> TRUE FALSE 1.378977e+00
## 217 <NA> TRUE FALSE 1.376050e+00
## 354 <NA> TRUE FALSE 1.356698e+00
## 104 <NA> TRUE FALSE 1.356510e+00
## 110 <NA> TRUE FALSE 1.347434e+00
## 235 <NA> TRUE FALSE 1.345955e+00
## 6 <NA> TRUE FALSE 1.337921e+00
## 373 <NA> TRUE FALSE 1.334523e+00
## 228 <NA> TRUE FALSE 1.324313e+00
## 27 <NA> TRUE FALSE 1.282086e+00
## 295 <NA> TRUE TRUE 1.277545e+00
## 193 <NA> TRUE FALSE 1.276831e+00
## 70 <NA> TRUE FALSE 1.276356e+00
## 315 <NA> TRUE TRUE 1.256679e+00
## 302 <NA> TRUE TRUE 1.242592e+00
## 299 <NA> TRUE FALSE 1.239209e+00
## 178 <NA> TRUE FALSE 1.234727e+00
## 12 <NA> TRUE TRUE 1.231878e+00
## 19 <NA> TRUE FALSE 1.199858e+00
## 106 <NA> TRUE TRUE 1.173028e+00
## 7 <NA> TRUE FALSE 1.161884e+00
## 85 <NA> TRUE FALSE 1.161036e+00
## 282 <NA> TRUE FALSE 1.159930e+00
## 200 <NA> TRUE FALSE 1.157188e+00
## 111 <NA> TRUE FALSE 1.156313e+00
## 113 <NA> TRUE FALSE 1.128949e+00
## 332 <NA> TRUE TRUE 1.120115e+00
## 368 <NA> TRUE FALSE 1.106651e+00
## 89 <NA> TRUE TRUE 1.095575e+00
## 231 <NA> TRUE FALSE 1.077000e+00
## 102 <NA> TRUE FALSE 1.072018e+00
## 264 <NA> TRUE FALSE 1.071941e+00
## 353 <NA> TRUE FALSE 1.068504e+00
## 154 <NA> TRUE FALSE 1.065379e+00
## 40 <NA> TRUE TRUE 1.052985e+00
## 277 <NA> TRUE FALSE 1.047051e+00
## 117 <NA> TRUE FALSE 1.045969e+00
## 246 <NA> TRUE TRUE 1.041244e+00
## 185 <NA> TRUE FALSE 1.036776e+00
## 271 <NA> TRUE FALSE 1.027185e+00
## 76 <NA> TRUE TRUE 9.997333e-01
## 325 <NA> TRUE FALSE 9.928496e-01
## 77 <NA> TRUE FALSE 9.919446e-01
## 139 <NA> TRUE FALSE 9.838134e-01
## 4 <NA> TRUE FALSE 9.769320e-01
## 269 <NA> TRUE FALSE 9.705752e-01
## 100 <NA> TRUE FALSE 9.672511e-01
## 8 <NA> TRUE TRUE 9.667640e-01
## 34 <NA> TRUE TRUE 9.597015e-01
## 141 <NA> TRUE TRUE 9.589176e-01
## 97 <NA> TRUE FALSE 9.542522e-01
## 213 <NA> TRUE FALSE 9.521729e-01
## 58 <NA> TRUE TRUE 9.382525e-01
## 56 <NA> TRUE FALSE 9.346640e-01
## 309 <NA> TRUE FALSE 9.341896e-01
## 199 <NA> TRUE FALSE 9.188157e-01
## 293 <NA> TRUE TRUE 9.145557e-01
## 20 <NA> TRUE FALSE 9.136048e-01
## 53 <NA> TRUE FALSE 9.132817e-01
## 215 <NA> TRUE FALSE 9.110084e-01
## 364 <NA> TRUE FALSE 9.017313e-01
## 3 <NA> TRUE FALSE 8.985659e-01
## 103 <NA> TRUE FALSE 8.865934e-01
## 109 <NA> TRUE FALSE 8.642038e-01
## 288 <NA> TRUE TRUE 8.600769e-01
## 339 <NA> TRUE FALSE 8.590769e-01
## 203 <NA> TRUE TRUE 8.351035e-01
## 265 <NA> TRUE FALSE 8.331659e-01
## 11 <NA> TRUE TRUE 8.331155e-01
## 30 <NA> TRUE TRUE 8.284208e-01
## 153 <NA> TRUE FALSE 8.248325e-01
## 243 A.life TRUE FALSE 8.145283e-01
## 119 <NA> TRUE FALSE 8.059668e-01
## 230 <NA> TRUE FALSE 8.042957e-01
## 241 <NA> TRUE FALSE 8.013139e-01
## 207 <NA> TRUE FALSE 8.009486e-01
## 319 <NA> TRUE FALSE 7.980231e-01
## 247 <NA> TRUE FALSE 7.944751e-01
## 145 <NA> TRUE FALSE 7.819763e-01
## 42 <NA> TRUE TRUE 7.797478e-01
## 22 <NA> TRUE FALSE 7.792762e-01
## 201 <NA> TRUE FALSE 7.785761e-01
## 329 <NA> TRUE FALSE 7.683640e-01
## 284 <NA> TRUE TRUE 7.679343e-01
## 38 <NA> TRUE FALSE 7.553197e-01
## 79 <NA> TRUE FALSE 7.540194e-01
## 120 <NA> TRUE FALSE 7.518264e-01
## 136 <NA> TRUE FALSE 7.450761e-01
## 18 <NA> TRUE FALSE 7.425963e-01
## 148 <NA> TRUE FALSE 7.412264e-01
## 255 <NA> TRUE FALSE 7.300357e-01
## 296 <NA> TRUE TRUE 7.265489e-01
## 262 <NA> TRUE FALSE 7.253745e-01
## 252 <NA> TRUE TRUE 7.245200e-01
## 202 <NA> TRUE FALSE 7.222191e-01
## 286 <NA> TRUE FALSE 7.209228e-01
## 15 <NA> TRUE FALSE 7.195183e-01
## 147 <NA> TRUE FALSE 7.100707e-01
## 208 <NA> TRUE FALSE 7.090976e-01
## 352 <NA> TRUE FALSE 7.074339e-01
## 149 <NA> TRUE FALSE 7.047624e-01
## 251 <NA> TRUE TRUE 7.009180e-01
## 152 <NA> TRUE TRUE 7.000440e-01
## 198 <NA> TRUE FALSE 6.933892e-01
## 363 <NA> TRUE FALSE 6.927683e-01
## 338 <NA> TRUE FALSE 6.918521e-01
## 205 <NA> TRUE FALSE 6.879783e-01
## 86 <NA> TRUE TRUE 6.755294e-01
## 351 <NA> TRUE FALSE 6.685467e-01
## 55 <NA> TRUE TRUE 6.637768e-01
## 62 <NA> TRUE FALSE 6.633486e-01
## 371 <NA> TRUE TRUE 6.631579e-01
## 283 <NA> TRUE FALSE 6.619130e-01
## 165 <NA> TRUE FALSE 6.575080e-01
## 226 <NA> TRUE TRUE 6.537250e-01
## 5 <NA> TRUE TRUE 6.513645e-01
## 278 <NA> TRUE FALSE 6.504590e-01
## 238 <NA> TRUE FALSE 6.377919e-01
## 144 <NA> TRUE TRUE 6.291639e-01
## 167 <NA> TRUE TRUE 6.228465e-01
## 82 <NA> TRUE FALSE 6.227290e-01
## 16 <NA> TRUE TRUE 6.201442e-01
## 23 <NA> TRUE TRUE 6.182969e-01
## 237 <NA> TRUE FALSE 6.179448e-01
## 227 <NA> TRUE FALSE 6.168379e-01
## 29 <NA> TRUE FALSE 6.128681e-01
## 365 <NA> TRUE FALSE 6.032202e-01
## 10 <NA> TRUE FALSE 6.028949e-01
## 182 <NA> TRUE FALSE 6.018819e-01
## 122 <NA> TRUE FALSE 5.914403e-01
## 96 <NA> TRUE TRUE 5.797991e-01
## 171 <NA> TRUE TRUE 5.773829e-01
## 172 <NA> TRUE FALSE 5.708311e-01
## 240 <NA> TRUE TRUE 5.686034e-01
## 214 <NA> TRUE FALSE 5.632897e-01
## 242 <NA> TRUE TRUE 5.586069e-01
## 291 <NA> TRUE FALSE 5.573902e-01
## 75 <NA> TRUE TRUE 5.519335e-01
## 35 <NA> TRUE TRUE 5.423197e-01
## 60 <NA> TRUE FALSE 5.227290e-01
## 334 <NA> TRUE FALSE 5.193850e-01
## 142 <NA> TRUE FALSE 5.150206e-01
## 159 <NA> TRUE TRUE 5.137620e-01
## 335 <NA> TRUE TRUE 5.100376e-01
## 219 <NA> TRUE TRUE 5.083006e-01
## 336 <NA> TRUE FALSE 5.046609e-01
## 273 <NA> TRUE FALSE 5.026711e-01
## 179 <NA> TRUE TRUE 4.965534e-01
## 13 <NA> TRUE FALSE 4.950049e-01
## 107 <NA> TRUE FALSE 4.945070e-01
## 320 <NA> TRUE FALSE 4.786928e-01
## 163 <NA> TRUE TRUE 4.713294e-01
## 331 <NA> TRUE FALSE 4.690384e-01
## 164 <NA> TRUE FALSE 4.689703e-01
## 326 A.nausea TRUE FALSE 4.681225e-01
## 239 <NA> TRUE FALSE 4.659589e-01
## 328 <NA> TRUE FALSE 4.609521e-01
## 330 <NA> TRUE TRUE 4.470171e-01
## 32 <NA> TRUE TRUE 4.440060e-01
## 41 <NA> TRUE FALSE 4.424712e-01
## 99 <NA> TRUE FALSE 4.404584e-01
## 84 <NA> TRUE FALSE 4.402143e-01
## 197 <NA> TRUE TRUE 4.386764e-01
## 317 <NA> TRUE TRUE 4.356813e-01
## 150 <NA> TRUE TRUE 4.333605e-01
## 155 <NA> TRUE FALSE 4.259778e-01
## 95 A.point TRUE FALSE 4.227088e-01
## 50 <NA> TRUE FALSE 4.162381e-01
## 298 <NA> TRUE TRUE 4.154826e-01
## 69 <NA> TRUE FALSE 4.100121e-01
## 156 <NA> TRUE TRUE 4.074903e-01
## 186 <NA> TRUE FALSE 4.069902e-01
## 72 <NA> TRUE TRUE 4.040278e-01
## 81 <NA> TRUE TRUE 3.965309e-01
## 166 <NA> TRUE FALSE 3.957035e-01
## 132 <NA> TRUE TRUE 3.927318e-01
## 125 <NA> TRUE FALSE 3.905017e-01
## 157 <NA> TRUE FALSE 3.881505e-01
## 276 <NA> TRUE FALSE 3.876879e-01
## 274 <NA> TRUE FALSE 3.874571e-01
## 174 <NA> TRUE FALSE 3.851609e-01
## 54 <NA> TRUE TRUE 3.813001e-01
## 224 <NA> TRUE FALSE 3.811110e-01
## 220 <NA> TRUE TRUE 3.791531e-01
## 130 <NA> TRUE TRUE 3.786759e-01
## 94 <NA> TRUE FALSE 3.699273e-01
## 31 <NA> TRUE TRUE 3.636233e-01
## 221 <NA> TRUE FALSE 3.580076e-01
## 137 <NA> TRUE TRUE 3.567457e-01
## 294 <NA> TRUE TRUE 3.516381e-01
## 73 <NA> TRUE FALSE 3.501763e-01
## 311 <NA> TRUE FALSE 3.501269e-01
## 192 <NA> TRUE TRUE 3.479555e-01
## 134 <NA> TRUE TRUE 3.447064e-01
## 59 <NA> TRUE TRUE 3.400594e-01
## 190 <NA> TRUE FALSE 3.397929e-01
## 115 <NA> TRUE TRUE 3.395485e-01
## 2 <NA> TRUE TRUE 3.387895e-01
## 324 <NA> TRUE FALSE 3.355332e-01
## 112 <NA> TRUE FALSE 3.307872e-01
## 151 <NA> TRUE TRUE 3.217259e-01
## 249 <NA> TRUE TRUE 3.181415e-01
## 108 <NA> TRUE FALSE 3.169618e-01
## 168 <NA> TRUE FALSE 3.166005e-01
## 52 <NA> TRUE TRUE 3.128434e-01
## 287 <NA> TRUE TRUE 3.101846e-01
## 126 <NA> TRUE TRUE 3.101823e-01
## 187 <NA> TRUE FALSE 3.097861e-01
## 173 <NA> TRUE FALSE 3.083751e-01
## 356 <NA> TRUE FALSE 3.017310e-01
## 234 <NA> TRUE TRUE 2.934947e-01
## 301 <NA> TRUE TRUE 2.888592e-01
## 158 <NA> TRUE TRUE 2.875472e-01
## 98 <NA> TRUE FALSE 2.846076e-01
## 189 <NA> TRUE TRUE 2.845946e-01
## 261 <NA> TRUE FALSE 2.843712e-01
## 218 <NA> TRUE FALSE 2.767329e-01
## 289 <NA> TRUE FALSE 2.746044e-01
## 128 <NA> TRUE TRUE 2.586523e-01
## 57 <NA> TRUE TRUE 2.553101e-01
## 260 <NA> TRUE FALSE 2.332121e-01
## 188 <NA> TRUE FALSE 2.306047e-01
## 170 <NA> TRUE FALSE 2.303897e-01
## 21 <NA> TRUE TRUE 2.297166e-01
## 232 <NA> TRUE FALSE 2.282553e-01
## 307 <NA> TRUE FALSE 2.255300e-01
## 92 <NA> TRUE FALSE 2.179265e-01
## 281 <NA> TRUE TRUE 2.021689e-01
## 263 <NA> TRUE TRUE 2.008108e-01
## 191 <NA> TRUE TRUE 1.944081e-01
## 25 <NA> TRUE TRUE 1.937381e-01
## 258 <NA> TRUE TRUE 1.756580e-01
## 210 <NA> TRUE TRUE 1.727850e-01
## 350 <NA> TRUE FALSE 1.698541e-01
## 169 <NA> TRUE TRUE 1.668714e-01
## 270 <NA> TRUE FALSE 1.612892e-01
## 87 <NA> TRUE TRUE 1.533848e-01
## 101 <NA> TRUE TRUE 1.524364e-01
## 135 <NA> TRUE FALSE 1.515588e-01
## 290 <NA> TRUE FALSE 1.500599e-01
## 279 <NA> TRUE TRUE 1.496071e-01
## 212 <NA> TRUE TRUE 1.440721e-01
## 121 <NA> TRUE FALSE 1.435747e-01
## 209 <NA> TRUE FALSE 1.347386e-01
## 162 <NA> TRUE TRUE 1.277322e-01
## 138 <NA> TRUE TRUE 1.247976e-01
## 233 <NA> TRUE TRUE 1.191385e-01
## 74 <NA> TRUE FALSE 1.149744e-01
## 256 <NA> TRUE TRUE 1.116638e-01
## 123 <NA> TRUE TRUE 1.089788e-01
## 280 <NA> TRUE FALSE 9.470030e-02
## 272 <NA> TRUE FALSE 9.337679e-02
## 45 <NA> TRUE TRUE 3.384106e-02
## 184 <NA> TRUE TRUE 3.206004e-02
## 71 <NA> TRUE TRUE 1.733491e-03
## 131 <NA> TRUE FALSE 1.597501e-03
## 322 <NA> TRUE FALSE 0.000000e+00
## 1 <NA> TRUE FALSE NA
## 129 <NA> FALSE TRUE NA
## 355 <NA> FALSE NA NA
## 376 <NA> NA FALSE NA# Used again in predict.data.new chunk
glb_analytics_diag_plots <- function(obs_df) {
if (length(vars <- subset(glb_feats_df, importance > 0)$id) > 5) {
warning("Limiting important feature scatter plots to 5 out of ", length(vars))
vars <- vars[1:5]
}
for (var in vars) {
plot_df <- melt(obs_df, id.vars=var,
measure.vars=c(glb_rsp_var, glb_rsp_var_out))
# if (var == "<feat_name>") print(myplot_scatter(plot_df, var, "value",
# facet_colcol_name="variable") +
# geom_vline(xintercept=<divider_val>, linetype="dotted")) else
print(myplot_scatter(plot_df, var, "value", colorcol_name="variable",
facet_colcol_name="variable", jitter=TRUE) +
guides(color=FALSE))
}
if (glb_is_regression) {
# plot_vars_df <- subset(glb_feats_df, importance >
# glb_feats_df[glb_feats_df$id == ".rnorm", "importance"])
plot_vars_df <- orderBy(~ -importance, glb_feats_df)
if (nrow(plot_vars_df) == 0)
warning("No important features in glb_fin_mdl") else
print(myplot_prediction_regression(df=obs_df,
feat_x=ifelse(nrow(plot_vars_df) > 1, plot_vars_df$id[2],
".rownames"),
feat_y=plot_vars_df$id[1],
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
id_vars=glb_id_vars)
# + facet_wrap(reformulate(plot_vars_df$id[2])) # if [1 or 2] is a factor
# + geom_point(aes_string(color="<col_name>.fctr")) # to color the plot
)
}
if (glb_is_classification) {
if (nrow(plot_vars_df <- subset(glb_feats_df, importance > 0)) == 0)
warning("No features in selected model are statistically important")
else print(myplot_prediction_classification(df=obs_df,
feat_x=ifelse(nrow(plot_vars_df) > 1, plot_vars_df$id[2],
".rownames"),
feat_y=plot_vars_df$id[1],
rsp_var=glb_rsp_var,
rsp_var_out=glb_rsp_var_out,
id_vars=glb_id_vars)
# + geom_hline(yintercept=<divider_val>, linetype = "dotted")
)
}
}
glb_analytics_diag_plots(obs_df=glb_trnent_df)## Warning in glb_analytics_diag_plots(obs_df = glb_trnent_df): Limiting
## important feature scatter plots to 5 out of 371
## title
## 1 Treatment of Hodgkin's disease and other cancers with 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU; NSC-409962).
## 2 Mometasone furoate effect on acute skin toxicity in breast cancer patients receiving radiotherapy: a phase III double-blind, randomized trial from the North Central Cancer Treatment Group N06C4.
## 1824 Toxicity of older and younger patients treated with adjuvant chemotherapy for node-positive breast cancer: the Cancer and Leukemia Group B Experience.
## abstract
## 1
## 2 PURPOSE: A two-arm, double-blind, randomized trial was performed to evaluate the effect of 0.1% mometasone furoate (MMF) on acute skin-related toxicity in patients undergoing breast or chest wall radiotherapy.METHODS AND MATERIALS: Patients with ductal carcinoma in situ or invasive breast carcinoma who were undergoing external beam radiotherapy to the breast or chest wall were randomly assigned to apply 0.1% MMF or placebo cream daily. The primary study endpoint was the provider-assessed maximal grade of Common Terminology Criteria for Adverse Events, version 3.0, radiation dermatitis. The secondary endpoints included provider-assessed Common Terminology Criteria for Adverse Events Grade 3 or greater radiation dermatitis and adverse event monitoring. The patient-reported outcome measures included the Skindex-16, the Skin Toxicity Assessment Tool, a Symptom Experience Diary, and a quality-of-life self-assessment. An assessment was performed at baseline, weekly during radiotherapy, and for 2 weeks after radiotherapy.RESULTS: A total of 176 patients were enrolled between September 21, 2007, and December 7, 2007. The provider-assessed primary endpoint showed no difference in the mean maximum grade of radiation dermatitis by treatment arm (1.2 for MMF vs. 1.3 for placebo; p = .18). Common Terminology Criteria for Adverse Events toxicity was greater in the placebo group (p = .04), primarily from pruritus. For the patient-reported outcome measures, the maximum Skindex-16 score for the MMF group showed less itching (p = .008), less irritation (p = .01), less symptom persistence or recurrence (p = .02), and less annoyance with skin problems (p = .04). The group's maximal Skin Toxicity Assessment Tool score showed less burning sensation (p = .02) and less itching (p = .002).CONCLUSION: Patients receiving daily MMF during radiotherapy might experience reduced acute skin toxicity compared with patients receiving placebo.
## 1824 PURPOSE: Older node-positive patients treated with newer adjuvant chemotherapy regimens have improvements in relapse-free and overall survival similar to younger patients. We compared toxicity of older and younger patients in three randomized trials of adjuvant chemotherapy. PATIENTS AND METHODS: Toxicity data were available for 93% of 6,642 patients enrolled. The three trials included: Cancer and Leukemia Group B (CALGB) 8541, a comparison of cyclophosphamide, doxorubicin, and fluorouracil in three dose schedules; CALGB 9344: cyclophosphamide and doxorubicin with or without paclitaxel; and CALGB 9741: cyclophosphamide, doxorubicin, and paclitaxel every 2 versus every 3 weeks. National Cancer Institute grade 3 to 5 toxicities were compared among age groups. RESULTS: Seven percent of patients (n = 458) were age 65 or older, 3% were 70 or older, 38% were 51 to 64, and 55% were 50 or younger. Twenty-four deaths (0.4%) were attributed to treatment; seven (1.5%) of 486 in patients 65 or older, 10 (0.40%) of 2,480 in patients who were 51 to 64 years, and seven (0.19%) of 3,676 occurred in patients younger than 50. In multivariate analysis, older patients were significantly more likely to have grade 4 hematologic toxicity, to have discontinued treatment for toxicity, or to have died of acute myeloid leukemia/myelodysplastic syndrome. There were no significant differences in grade 3 to 4 nonhematologic toxicity. CONCLUSION: Healthy older patients who met the strict eligibility criteria for these trials had a higher rate of hematologic toxicity and treatment-related deaths than younger patients, but no increase in nonhematologic toxicity. Elderly patients treated with newer adjuvant chemotherapy regimens derive the same benefits from newer chemotherapy regimens as younger patients but should be cautioned about the increased risk of toxicity and treatment-related death.
## trial .rnorm trial.fctr T.adjuv T.advanc T.breast T.cancer
## 1 1 -1.7487442 Y 0 0 0 1
## 2 0 0.6403752 N 0 0 1 2
## 1824 0 -1.1594386 N 1 0 1 2
## T.chemotherapi T.clinic T.combin T.compar T.cyclophosphamid
## 1 0 0 0 0 0
## 2 0 0 0 0 0
## 1824 1 0 0 0 0
## T.docetaxel T.doxorubicin T.earli T.effect T.group T.iii T.metastat
## 1 0 0 0 0 0 0 0
## 2 0 0 0 1 1 1 0
## 1824 0 0 0 0 1 0 0
## T.patient T.phase T.plus T.postmenopaus T.random T.randomis T.respons
## 1 0 0 0 0 0 0 0
## 2 1 1 0 0 1 0 0
## 1824 1 0 0 0 0 0 0
## T.result T.studi T.tamoxifen T.therapi T.treatment T.trial T.versus
## 1 0 0 0 0 1 0 0
## 2 0 0 0 0 1 1 0
## 1824 0 0 0 0 0 0 0
## T.women T.has.http T.num.chars T.num.words T.num.words.unq A.X0001
## 1 0 0 110 7 7 0
## 2 0 0 194 22 21 0
## 1824 0 0 150 14 13 0
## A.X001 A.X005 A.X100 A.X500 A.X5fluorouracil A.accord A.achiev
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 0 0
## A.activ A.addit A.adjuv A.administ A.administr A.advanc A.advers
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 4
## 1824 0 0 3 0 0 0 0
## A.age A.agent A.aim A.alon A.also A.although A.among A.analys
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0
## 1824 2 0 0 0 0 0 1 0
## A.analysi A.analyz A.andor A.anthracyclin A.appear A.arm A.aromatas
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 1 0
## 1824 1 0 0 0 0 0 0
## A.assess A.assign A.associ A.avail A.axillari A.background A.base
## 1 0 0 0 0 0 0 0
## 2 3 1 0 0 0 0 0
## 1824 0 0 0 1 0 0 0
## A.baselin A.benefit A.better A.bone A.breast A.can A.cancer
## 1 0 0 0 0 0 0 0
## 2 1 0 0 0 3 0 0
## 1824 0 1 0 0 0 0 2
## A.carcinoma A.case A.caus A.cell A.chang A.characterist
## 1 0 0 0 0 0 0
## 2 2 0 0 0 0 0
## 1824 0 0 0 0 0 0
## A.chemotherapi A.clinic A.cmf A.combin A.common A.compar A.comparison
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 3 1 0
## 1824 4 0 0 0 0 2 1
## A.complet A.conclus A.conduct A.confid A.confirm A.consid A.consist
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 1 0 0 0 0 0
## A.continu A.control A.correl A.cours A.cycl A.cyclophosphamid A.daili
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 2
## 1824 0 0 0 0 0 3 0
## A.data A.day A.death A.decreas A.defin A.demonstr A.design A.detect
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0
## 1824 1 0 3 0 0 0 0 0
## A.determin A.develop A.dfs A.differ A.diseas A.diseasefre A.distant
## 1 0 0 0 0 0 0 0
## 2 0 0 0 1 0 0 0
## 1824 0 0 0 1 0 0 0
## A.docetaxel A.dose A.doubleblind A.doxorubicin A.drug A.due A.durat
## 1 0 0 0 0 0 0 0
## 2 0 0 1 0 0 0 0
## 1824 0 1 0 3 0 0 0
## A.earli A.effect A.efficaci A.eight A.either A.elig A.end A.endocrin
## 1 0 0 0 0 0 0 0 0
## 2 0 1 0 0 0 0 0 0
## 1824 0 0 0 0 0 1 0 0
## A.endpoint A.enrol A.enter A.epirubicin A.estim A.estrogen A.evalu
## 1 0 0 0 0 0 0 0
## 2 3 1 0 0 0 0 1
## 1824 0 1 0 0 0 0 0
## A.event A.everi A.evid A.examin A.experienc A.express A.factor
## 1 0 0 0 0 0 0 0
## 2 4 0 0 0 0 0 0
## 1824 0 2 0 0 0 0 0
## A.failur A.find A.first A.firstlin A.five A.fluorouracil A.follow
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 1 0
## A.followup A.found A.four A.frequent A.function. A.general A.given
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 0 0
## A.grade A.greater A.group A.growth A.hazard A.hematolog A.her2 A.high
## 1 0 0 0 0 0 0 0 0
## 2 3 2 3 0 0 0 0 0
## 1824 3 0 2 0 0 2 0 0
## A.higher A.histolog A.hormon A.howev A.human A.hundr A.identifi A.iii
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0
## 1824 1 0 0 0 0 0 0 0
## A.import A.improv A.incid A.includ A.increas A.independ A.indic
## 1 0 0 0 0 0 0 0
## 2 0 0 0 2 0 0 0
## 1824 0 1 0 1 2 0 0
## A.inform A.infus A.inhibitor A.initi A.interv A.intraven A.investig
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 0 0
## A.involv A.larg A.least A.less A.level A.life A.limit A.local
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 6 0 0 0 0
## 1824 0 0 0 0 0 0 0 0
## A.longer A.low A.lower A.lymph A.main A.major A.marker A.mastectomi
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 0 0 0
## A.may A.mbc A.mean A.measur A.median A.metastas A.metastat A.method
## 1 0 0 0 0 0 0 0 0
## 2 0 0 1 2 0 0 0 0
## 1824 0 0 0 0 0 0 0 1
## A.methotrex A.mgm2 A.model A.month A.multicent A.multivari A.nausea
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 1 0
## A.need A.negat A.neoadjuv A.neutropenia A.new A.node A.nodeposit
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 0 1
## A.number A.object A.observ A.obtain A.occur A.one A.oper A.oral
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0
## 1824 0 0 0 0 1 0 0 0
## A.outcom A.overal A.paclitaxel A.partial A.particip A.patholog
## 1 0 0 0 0 0 0
## 2 2 0 0 0 0 0
## 1824 0 1 2 0 0 0
## A.patient A.per A.perform A.period A.phase A.placebo A.plus A.point
## 1 0 0 0 0 0 0 0 0
## 2 5 0 2 0 0 4 0 0
## 1824 14 0 0 0 0 0 0 0
## A.popul A.posit A.possibl A.postmenopaus A.postop A.potenti A.predict
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 0 0 0 0
## A.premenopaus A.present A.pretreat A.prevent A.previous A.primari
## 1 0 0 0 0 0 0
## 2 0 0 0 0 0 2
## 1824 0 0 0 0 0 0
## A.prior A.profil A.progesteron A.prognost A.progress A.progressionfre
## 1 0 0 0 0 0 0
## 2 0 0 0 0 0 0
## 1824 0 0 0 0 0 0
## A.prolong A.proport A.prospect A.provid A.purpos A.qualiti
## 1 0 0 0 0 0 0
## 2 0 0 0 0 1 0
## 1824 0 0 0 0 1 0
## A.radiotherapi A.random A.randomis A.rang A.rate A.ratio A.receiv
## 1 0 0 0 0 0 0 0
## 2 3 2 0 0 0 0 2
## 1824 0 1 0 0 1 0 0
## A.receptor A.receptorposit A.recurr A.reduc A.reduct A.regard
## 1 0 0 0 0 0 0
## 2 0 0 1 1 0 0
## 1824 0 0 0 0 0 0
## A.regimen A.regress A.relaps A.relat A.remain A.report A.requir
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 3 0 0 0 0 0 0
## A.respect A.respond A.respons A.result A.risk A.safeti A.sampl
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 0 1 1 0 0
## A.schedul A.score A.second A.secondari A.seen A.select A.sequenti
## 1 0 0 0 0 0 0 0
## 2 0 2 0 1 0 0 0
## 1824 1 0 0 0 0 0 0
## A.serum A.set A.seven A.sever A.show A.shown A.side A.signific
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 3 0 0 0
## 1824 0 0 3 0 0 0 0 2
## A.similar A.singl A.site A.six A.size A.stabl A.stage A.standard
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0 0
## 1824 1 0 0 0 0 0 0 0
## A.start A.statist A.status A.studi A.subgroup A.suggest A.superior
## 1 0 0 0 0 0 0 0
## 2 0 0 0 1 0 0 0
## 1824 0 0 0 0 0 0 0
## A.support A.surgeri A.surviv A.system A.tamoxifen A.term A.test
## 1 0 0 0 0 0 0 0
## 2 0 0 0 0 0 0 0
## 1824 0 0 1 0 0 0 0
## A.therapi A.three A.time A.tissu A.toler A.total A.toxic A.treat
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 0 0 1 5 0
## 1824 0 3 0 0 0 0 9 2
## A.treatment A.trend A.trial A.tumor A.tumour A.two A.type A.use
## 1 0 0 0 0 0 0 0 0
## 2 1 0 1 0 0 0 0 0
## 1824 2 0 3 0 0 0 0 0
## A.valu A.versus A.vomit A.week A.well A.wherea A.whether A.within
## 1 0 0 0 0 0 0 0 0
## 2 0 0 0 2 0 0 0 0
## 1824 0 1 0 1 0 0 0 0
## A.without A.women A.year A.has.http A.num.chars A.num.words
## 1 0 0 0 0 0 0
## 2 0 0 0 0 1942 177
## 1824 1 0 1 0 1892 162
## A.num.words.unq trial.fctr.predict.Final.rf.prob
## 1 0 0.656
## 2 97 0.198
## 1824 92 0.194
## trial.fctr.predict.Final.rf trial.fctr.predict.Final.rf.accurate
## 1 Y TRUE
## 2 N TRUE
## 1824 N TRUE
## .label
## 1 .1
## 2 .2
## 1824 .1824
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all.prediction","model.final")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
## 3.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: data.training.all.prediction
## 4.0000 5 0 1 1 1
## 4.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: model.final
## 5.0000 4 0 0 2 1
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="predict.data.new",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed12 fit.data.training.all 6 1 511.814
## elapsed13 predict.data.new 7 0 517.551Step 7: predict data.new
# Compute final model predictions
glb_newent_df <- glb_get_predictions(glb_newent_df)
glb_analytics_diag_plots(obs_df=glb_newent_df)## Warning in glb_analytics_diag_plots(obs_df = glb_newent_df): Limiting
## important feature scatter plots to 5 out of 371
## title
## 6 Prognostic value of extracapsular tumor spread for locoregional control in premenopausal patients with node-positive breast cancer treated with classical cyclophosphamide, methotrexate, and fluorouracil: long-term observations from International Breast Cancer Study Group Trial VI.
## 18 Second malignancies after CMF for resectable breast cancer.
## 21 Her-2/neu overexpression and response to oophorectomy plus tamoxifen adjuvant therapy in estrogen receptor-positive premenopausal women with operable breast cancer.
## 26 Tetracosactrin vs. methylprednisolone in the prevention of emesis in patients receiving FEC regimen for breast cancer.
## 31 Randomized phase III trial of adjuvant epirubicin followed by cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) versus CMF followed by epirubicin in patients with node-negative or 1-3 node-positive rapidly proliferating breast cancer.
## 90 Weekly low-dose doxorubicin with or without high-dose medroxyprogesterone acetate as secondary treatment in metastatic breast cancer--a randomized trial.
## 98 Good clinical response of breast cancers to neoadjuvant chemoendocrine therapy is associated with improved overall survival.
## 102 Dose intensity in cancer chemotherapy.
## 169 Phase I study of L-asparaginase (NSC 109229).
## 187 Double-blind, placebo-controlled, randomised phase II trial of IH636 grape seed proanthocyanidin extract (GSPE) in patients with radiation-induced breast induration.
## 532 Serum methionine depletion without side effects by methioninase in metastatic breast cancer patients.
## 677 A phase I/II trial of a polysaccharide extract from Grifola frondosa (Maitake mushroom) in breast cancer patients: immunological effects.
## 743 Classical markers like ER and ki-67, but also survivin and pERK, could be involved in the pathological response to gemcitabine, adriamycin and paclitaxel (GAT) in locally advanced breast cancer patients: results from the GEICAM/2002-01 phase II study.
## 750 Safety results from a phase III study (TURANDOT trial by CECOG) of first-line bevacizumab in combination with capecitabine or paclitaxel for HER-2-negative locally recurrent or metastatic breast cancer.
## 792 Intravenous or oral vinorelbine plus capecitabine as first-line treatment in HER2- metastatic breast cancer: joint analysis of 2 consecutive prospective phase II trials.
## 1832 Adjuvant chemotherapy for premenopausal breast cancer: a meta-analysis using quality-adjusted survival.
## abstract
## 6 PURPOSE: We sought to determine retrospectively whether extracapsular spread (ECS) might identify a subgroup that could benefit from radiotherapy after mastectomy, especially patients with 1 to 3 positive lymph nodes (LN1-3+). PATIENTS AND METHODS: We randomized 1,475 premenopausal women with node-positive breast cancer to three, six, or nine courses of "classical" CMF (cyclophosphamide, methotrexate, and fluorouracil). After a review of all pathology forms, 933 patients (63%) had information on the presence or absence of ECS. ECS was present in 49.5%. The median follow-up was 10 years. RESULTS: In univariate analyses, ECS was associated with worse disease-free survival (DFS) and overall survival (OS). In multivariate analyses adjusting for tumor size, vessel invasion, surgery type, and age group, ECS remained significant (DFS: hazard ratio, 1.61; 95% CI, 1.34 to 1.93; P < .0001; OS: 1.67; 95% CI, 1.34 to 2.08; P < .0001). However, ECS was not significant when the number of positive nodes was added. The locoregional failure rate +/- distant failure (LRF +/- distant failure) within 10 years was estimated at 19% (+/- 2%) without ECS, versus 27% (+/- 2%) with ECS. The difference was statistically significant in univariate analyses, but not after adjusting for the number of positive nodes. No independent effect of ECS on DFS, OS, or LRF could be confirmed within the subgroup of 382 patients with LN1-3+ treated with mastectomy without radiotherapy. CONCLUSION: Our results do not support an independent prognostic value of ECS, nor its use as an indication for irradiation in premenopausal patients with LN1-3+ treated with classical CMF. However, we could not examine whether extensive ECS is of prognostic importance.
## 18 From June 1973 to May 1978, a total of 845 women with resectable breast cancer and positive axillary nodes were entered into two consecutive randomized studies evaluating adjuvant chemotherapy. All patients were subjected to radical or modified radical mastectomy, none received postoperative radiation, and 666 were administered adjuvant CMF (cyclophosphamide, methotrexate, and fluorouracil). After a median follow-up in excess of 10 years, no cases of acute nonlymphocytic leukemia were detected, but 21 second solid tumors other than contralateral breast carcinoma were documented. The cumulative frequency was 4% +/- 1.9% after surgery alone, and 4.2% +/- 1.03% following adjuvant CMF. No differences were observed between patients aged up to 50 years (surgery, 3.1% +/- 2.2%; CMF, 3.3% +/- 1.3%) or older than 50 years (surgery, 4.5% +/- 2.6%; CMF, 5.2% +/- 1.8%). During the same period, a total of 29 contralateral breast carcinomas were documented for a cumulative frequency of 3.7% +/- 1.7% after surgery alone and of 5.2% +/- 1.4% following adjuvant CMF, respectively. We conclude that, at present, there is no evidence for an increased risk of second malignancies following adjuvant CMF as given in this series. Our findings would suggest that second tumors documented so far cannot be entirely ascribed to treatment with adjuvant chemotherapy, but they could be due to a chance association.
## 21 PURPOSE: Studies evaluating the relationship of HER-2/neu breast tumor status and response to adjuvant endocrine therapy have reached conflicting conclusions about resistance of HER-2/neu-positive tumors to this treatment. We studied 282 patients participating in a randomized controlled trial of adjuvant oophorectomy and tamoxifen or observation who had estrogen receptor-positive tumors and whose tumors were evaluated for HER-2/neu overexpression by immunohistochemistry. PATIENTS AND METHODS: Univariate and multivariate Cox proportional hazards regression models and Kaplan-Meier disease-free and overall survival estimate methods were used. RESULTS: HER-2/neu overexpression was a negative prognostic factor for overall survival. In univariate analyses, in HER-2/neu-positive patients, the hazard ratio (HR) for disease-free survival (DFS) with adjuvant endocrine therapy was 0.37 (95% confidence interval [CI], 0.26 to 0.89); for HER-2/neu-negative patients, the corresponding HR for DFS was 0.48 (95% CI, 0.31 to 0.71). The overall survival (OS) data were HR=0.26 (95% CI, 0.07 to 0.92) and HR=0.68 (95% CI, 0.32 to 1.42) for HER-2/neu-positive and HER-2/neu-negative patients, respectively. In multivariate models, the P values for tests of interaction of HER-2/neu status and response to adjuvant endocrine therapy were 0.18 and 0.07 for DFS and OS, respectively. Kaplan-Meier DFS and OS curves and 3-year DFS estimates were consistent in showing greater benefit to the HER-2/neu-positive subgroup given adjuvant treatment. CONCLUSION: HER-2/neu overexpression does not adversely and may favorably influence response to adjuvant oophorectomy and tamoxifen treatment in patients with estrogen receptor-positive tumors.
## 26 0.5 mg tetracosactrin is considered to be equivalent to 40 mg methylprednisolone with regard to the induced cortisol secretion. 97 female breast cancer patients who received their first two FEC courses (epirubicin 50-75 mg/m2, 5-fluorouracil 500 mg/m2, cyclophosphamide 500 mg/m2) entered this randomised crossover study (76 had previously received an adjuvant treatment); tetracosactrin was administered intramuscularly and methylprednisolone intravenously immediately before chemotherapy administration. The tolerability was evaluated using a diary card during 5 days and patients were asked for their preference at the end of the two cycles. There was no difference either for vomiting (dry heaves were included) or nausea between the two treatments (the analysis was performed on day 1, the worse day of days 2 and 3 and the worse day of days 4 and 5). At day 1, 49% of the patients experienced no or mild nausea after tetracosactrin and 62% after methylprednisolone (not significant) (first period analysis); a complete control of vomiting (including dry heaves) was observed in 49% of the patients after tetracosactrin and 53% after methylprednisolone (not significant). No difference was observed between patients with or without previous chemotherapy. However, slightly more patients preferred tetracosactrin (P = 0.048).
## 31 Adjuvant cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) have proven highly effective in rapidly proliferating breast cancer (RPBC). It has also been seen that sequential administration of doxorubicin and CMF is superior to their alternation, especially in indolent tumors. In a phase III study, we evaluated whether adjuvant epirubicin (E) followed by CMF is superior to the inverse sequence in RPBC. Patients with node-negative or 1-3 node-positive RPBC (Thymidine Labeling Index > 3% or histological grade 3 or S-phase > 10% or Ki67 > 20%) were randomized to receive E (100 mg/m(2) i.v. d1, q21 days for 4 cycles) followed by CMF (600, 40, 600 mg/m(2) i.v. d1 and 8, q28 days for 4 cycles) (E → CMF) or CMF followed by E (CMF → E) or CMF for 6 cycles. From November 1997 to December 2004, 1066 patients were enrolled: E → CMF 440, CMF → E 438, and CMF 188. At a median follow-up of 69 months, 5-year OS was 91% (95% CI 88-94) for E → CMF and 93% (95% CI 90-95) for CMF → E, with adjusted hazard ratio of 0.88 (95% CI 0.58-1.35), and DFS was 80% in both arms, with adjusted hazard ratio of 0.99 (95% CI 0.73-1.33, Cox model). Adverse events were similar, apart from a higher rate of neutropenia in the CMF → E arm. No important differences in clinical outcome were observed between the two different sequences, making both a valid option in early breast cancer. Further molecular characterization of the tumors might help to identify subgroups achieving higher benefit from either sequence.
## 90
## 98 BACKGROUND: We present extended follow-up from a prospective randomised trial evaluating the role of neoadjuvant chemoendocrine therapy in the treatment of operable breast cancer. PATIENTS AND METHODS: 309 women were randomised to primary surgery followed by eight cycles of adjuvant mitoxantrone, methotrexate with tamoxifen (2MT) or 2MT with mitomycin-C (3MT) versus the same regimen for four cycles before followed by four cycles after surgery. For this analysis the median follow-up of patients was 112 months. RESULTS: After 10 years follow-up there is still no statistically significant difference in disease-free survival (DFS) (71% versus 71%) or overall survival (OS) (63% versus 70%) when comparing adjuvant versus neoadjuvant treatment, respectively. Of 144 evaluable patients in the neoadjuvant arm, 74 achieved a good clinical response and 70 patients achieved a poor clinical response. Good responders had a superior DFS (80% versus 64%, P=0.01) and OS (77% versus 63%, P=0.03) compared to poor responders. CONCLUSIONS: At 10 years, neoadjuvant and adjuvant treatment continue to have equivalent OS and DFS. Good clinical response to neoadjuvant chemotherapy is associated with superior DFS and OS. This supports the use of clinical response of primary breast cancer to neoadjuvant therapy as a surrogate marker of survival benefit.
## 102
## 169 22 patients received intravenously infused L-asparaginase (Escherichia coli) on a protocol for 5 weekly doses. 13 patients received 100 U/kg, 1 patient 500 U/kg, and 8 patients 1,500 U/kg. Only 3 of the 9 patients receiving 500 U/kg or more were able to complete the 5-week protocol. 11 of the 13 patients receiving 100 U/kg were able to complete the 5-week protocol. Significant tumor responses were not seen. CNS toxicity and allergic reactions were observed at high- and low-dose levels. There was no difference as to the degree of protein changes, BUN elevation, or liver function abnormalities at the different dose levels.
## 187 BACKGROUND AND PURPOSE: Tissue hardness (induration), pain and tenderness are common late adverse effects of curative radiotherapy for early breast cancer. The purpose of this study was to test the efficacy of IH636 grape seed proanthocyanidin extract (GSPE) in patients with tissue induration after high-dose radiotherapy for early breast cancer in a double-blind placebo-controlled randomised phase II trial. PATIENTS AND METHODS: Sixty-six eligible research volunteers with moderate or marked breast induration at a mean 10.8 years since radiotherapy for early breast cancer were randomised to active drug (n = 44) or placebo (n = 22). All patients were given grape seed proanthocyanidin extract (GSPE) 100 mg three times a day orally, or corresponding placebo capsules, for 6 months. The primary endpoint was percentage change in surface area (cm(2)) of palpable breast induration measured at the skin surface 12 months after randomisation. Secondary endpoints included change in photographic breast appearance and patient self-assessment of breast hardness, pain and tenderness. RESULTS: At 12 months post-randomisation, > or =50% reduction in surface area (cm(2)) of breast induration was recorded in 13/44 (29.5%) GSPE and 6/22 (27%) placebo group patients (NS). At 12 months post-randomisation, there was no significant difference between treatment and control groups in terms of external assessments of tissue hardness, breast appearance or patient self-assessments of breast hardness, pain or tenderness. CONCLUSIONS: The study failed to show efficacy of orally-administered GSPE in patients with breast induration following radiotherapy for breast cancer.
## 532 The growth dependence of human tumors on elevated levels of methionine has been shown in preclinical in vitro and in vivo studies to be a frequently occurring, highly effective, tumor-selective therapeutic target. High purity endotoxin-free methioninase was produced from Pseudomonas putida in order to develop anti-methionine chemotherapy targeting of human tumors. A pilot Phase I clinical trial has been initiated in order to determine methioninase toxicity, the pharmacokinetics of methioninase and methionine-depletion and maximum tolerated dose. A two hour i.v. infusion of 5,000 units (0.4 g) and 10,000 units (0.8 g) and a ten hour i.v. infusion of 20,000 units (1.6 g) of methioninase was administered to patient-1, patient-2, and patient-3, respectively. All patients had advanced breast cancer. Blood and urine samples were obtained at frequent intervals between 0 and 24 hours. The toxicity evaluations were carried out according to FDA criteria. Pharmacokinetics data were obtained for both methioninase and methionine levels in the serum. No acute clinical toxicity was observed for all the toxicity criteria measured in patient-1, patient-2 and patient-3. The depletion of serum methionine started within 30 minutes of the infusion, and was maintained for 4 hours after the infusion was completed in patient-1 and patient-2. The lowest serum methionine levels were 35% and 19% of the pretreatment level, respectively, in patient-1 and patient-2. Patient-3 received a ten hour i.v. infusion of 20,000 units of methioninase without any signs of side effects. Patient-3 maintained serum levels of methioninase as high as 50% of the maximum level for a subsequent 6 hours after infusion. Methionine was depleted over 200-fold from 23.1 microM to 0.1 microM by the 10-hour infusion of patient-3. No clinical toxicity was observed whatsoever in all the toxicity criteria measured in patient-3. The results of the methioninase pilot Phase 1 clinical trial suggested that i.v. infusion of the methioninase is safe and effectively depletes serum methionine without any signs of side effects. Clinical studies are continuing to determine the maximum length of time complete serum methionine depletion can be tolerated.
## 677 BACKGROUND: Cancer patients commonly use dietary supplements to "boost immune function". A polysaccharide extract from Grifola frondosa (Maitake extract) showed immunomodulatory effects in preclinical studies and therefore the potential for clinical use. Whether oral administration in human produces measurable immunologic effects, however, is unknown. METHODS: In a phase I/II dose escalation trial, 34 postmenopausal breast cancer patients, free of disease after initial treatment, were enrolled sequentially in five cohorts. Maitake liquid extract was taken orally at 0.1, 0.5, 1.5, 3, or 5 mg/kg twice daily for 3 weeks. Peripheral blood was collected at days -7, 0 (prior to the first dosing), 7, 14, and 21 for ex vivo analyses. The primary endpoints were safety and tolerability. RESULTS: No dose-limiting toxicity was encountered. Two patients withdrew prior to completion of the study due to grade I possibly related side effects: nausea and joint swelling in one patient; rash and pruritus in the second. There was a statistically significant association between Maitake and immunologic function (p < 0.0005). Increasing doses of Maitake increased some immunologic parameters and depressed others; the dose-response curves for many endpoints were non-monotonic with intermediate doses having either immune enhancing or immune suppressant effects compared with both high and low doses. CONCLUSIONS: Oral administration of a polysaccharide extract from Maitake mushroom is associated with both immunologically stimulatory and inhibitory measurable effects in peripheral blood. Cancer patients should be made aware of the fact that botanical agents produce more complex effects than assumed, and may depress as well as enhance immune function.
## 743 INTRODUCTION: The identification and validation of biomarkers of chemotherapy sensitivity is critical in order to individualise therapy in breast cancer. We evaluated pathological complete response (pCR) to GAT, and its correlation with tumour biomarkers before and after neoadjuvant chemotherapy. MATERIALS AND METHODS: Stage III (and stage II with T≥5 cm) breast cancer patients were included. Treatment consisted of adriamycin (40 mg/m(2)) day 1, and paclitaxel (150 mg/ m(2)) followed by gemcitabine (2000 mg/m(2)) day 2, every 14 days for six cycles. Tissue from pre-treatment biopsy and surgery was evaluated for biologic markers by immunohistochemistry. Two XPD single nucleotide polymorphisms (SNP) were also analysed. RESULTS: Forty-six patients entered the trial. Median age was 49.5 years (range 31-72); 25 patients (54%) were pre-menopausal; 12 (26%) were ER-PgR-negative; pCR was observed in 17% (95% CI: 6.4-28.4) of patients. Significant differences in marker expression (mean±SD) in correlation to pathological response were only found in Ki- 67. After treatment, tumours showed lower Ki-67-, surviving- and pERK-positive cells. No correlation between XPD polymorphisms and pCR was found. The overall response rate was 89% (95% CI: 80.1-98.1). Fifteen patients (33%) underwent breast-conserving surgery. The most frequent grade 3 or 4 toxicities were neutropenia (with one febrile neutropenia) and asthenia. CONCLUSION: These results show an effective regimen with acceptable tolerability. Our data suggest that not only classical markers (ER, Ki-67), but also survivin and pERK could be involved in the response to GAT, which may contribute to therapy individualisation in future study designs.
## 750 BACKGROUND: We report safety data from a randomised, phase III study (CECOG/BC.1.3.005) evaluating first-line bevacizumab plus paclitaxel or capecitabine for locally recurrent or metastatic breast cancer. PATIENTS AND METHODS: Patients aged ⩾18years with human epidermal growth factor receptor-2-negative breast adenocarcinoma were randomised to Arm A: bevacizumab 10mg/kg days 1 and 15; paclitaxel 90mg/m(2) days 1, 8, and 15, every 4weeks; or Arm B: bevacizumab 15mg/kg day 1; capecitabine 1000mg/m(2) b.i.d., days 1-14, every 3weeks, until disease progression, unacceptable toxicity or consent withdrawal. RESULTS: A post hoc interim safety analysis included 561 patients (Arm A: 284, Arm B: 277). The regimens demonstrated similar frequencies of all-grade and serious adverse events (SAEs), but different safety profiles. Treatment-related events occurred in 85.2% (Arm A) and 78.0% (Arm B) of patients. Fatigue was most common in Arm A (30.6% versus 23.5% Arm B), and hand-foot syndrome (HFS) most common in Arm B (49.5% versus 2.5% Arm A). Diarrhoea (Arm A: 0.4%, Arm B: 1.4%) and pulmonary embolism (Arm A: 0.7%, Arm B: 1.1%) were the most frequently reported SAEs. CONCLUSION: These findings are in-line with safety data for bevacizumab plus paclitaxel or capecitabine, reported in previous phase III trials.
## 792 BACKGROUND: The purpose of this study was to assess the activity and safety of the combination of vinorelbine (VNR) and capecitabine (CAP) as first-line treatment in HER2-negative (HER(-)) metastatic breast cancer (MBC). PATIENTS AND METHODS: Patients (42) enrolled in trial A received intravenous (i.v.) VNR 25 mg/m2 on days 1 and 8 of a 21-day cycle combined with CAP 1000 mg/m2 twice daily for 14 consecutive days followed by 1 week of rest. Trial B (46 patients) followed trial A when the oral formulation of VNR became available at our institution. Patients received oral VNR (60 mg/m(2) on days 1-8) combined with the same CAP schedule as in trial A. RESULTS: The response rate (RR) in trial A was 73.2% (95% confidence interval [CI], 56.4-82.8), including 12.2% complete responses (CRs). Clinical benefit was achieved in 78% of patients (95% CI, 63.2-87.9). In trial B, overall RR was 76% (95% CI, 62.0-86.0), with 13% CRs and clinical benefit of 80.4% (95% CI, 66.8-89.3). In trial A, median progression-free survival (PFS) was 8.2 months (range, 6-14+ months) and median overall survival (OS) was 32.4 months (range, 17-36+ months). In trial B, median PFS and OS were 8.8 months (range, 8-21+ months) and 34.3 months (14-39+ months), respectively. Treatment-related toxicity was manageable. Quality of life assessment showed a statistically significant difference regarding body image (p = .001), sexual functioning (p = .02), and future perspectives (p = .03) in women receiving chemotherapy fully by the oral route. CONCLUSION: This joint analysis shows that both tested schedules can produce high objective RRs with encouraging PFS, manageable toxicity profile, and suggested benefit on some aspects of quality of life for the fully oral combination.
## 1832 PURPOSE: Adjuvant chemotherapy for early breast cancer has been shown to offer an improvement in recurrence-free and overall survival, especially for younger women, but the acute toxic effects of this treatment discourage some physicians from prescribing it. The purpose of this analysis was to determine whether the benefit of 6 months of adjuvant CMF (cyclophosphamide, methotrexate, fluorouracil) treatment outweighs its costs in terms of toxic effects. METHODS: A meta-analysis of quality-adjusted survival was performed based on data from 1229 patients, aged 49 years or younger, randomized in eight trials comparing CMF versus no adjuvant systemic therapy. The eight trials were included in the worldwide overview conducted by the Early Breast Cancer Trialists' Collaborative Group. The Q-TWiST method was used in a meta-analysis that provided treatment comparisons incorporating differences in quality of life associated with the amount of time patients spend with subjective toxic effects, after relapse, and without symptoms of relapse. RESULTS: Within 6 years of follow-up evaluation for patients with node-positive disease, the benefit in terms of increased relapse-free and overall survival balanced the costs in terms of acute toxic side effects. This was true even for the extreme case in which a zero value was assigned to all 6 months during which patients might receive adjuvant CMF chemotherapy. Within 10 years of follow-up evaluation, treated patients gained an average of 1.5 years of relapse-free survival time, almost 1 year of overall survival time, and 1 year of time without symptoms and toxicity. CONCLUSIONS: Adjuvant chemotherapy for younger women with node-positive breast cancer provided substantial amounts of quality-adjusted survival time, even after accounting for costs associated with toxic effects of the treatment. The Q-TWiST method represents a valuable tool for comparing treatments because it incorporates patients' perceptions of their quality of life for therapeutic decision-making.
## trial .rnorm trial.fctr T.adjuv T.advanc T.breast T.cancer
## 6 0 -0.13338251 N 0 0 2 2
## 18 0 2.43143709 N 0 0 1 1
## 21 1 1.04906108 Y 1 0 1 1
## 26 0 1.10587666 N 0 0 1 1
## 31 1 -1.33467322 Y 1 0 1 1
## 90 1 0.62484238 Y 0 0 1 0
## 98 0 1.19838847 N 0 0 1 1
## 102 0 1.62210629 N 0 0 0 1
## 169 0 -0.72645834 N 0 0 0 0
## 187 0 0.16739095 N 0 0 1 0
## 532 0 -1.27306733 N 0 0 1 1
## 677 0 -0.63099642 N 0 0 1 1
## 743 0 -0.61939101 N 0 1 1 1
## 750 0 -1.86737032 N 0 0 1 1
## 792 0 -0.38147216 N 0 0 1 1
## 1832 0 -0.07397837 N 1 0 1 1
## T.chemotherapi T.clinic T.combin T.compar T.cyclophosphamid
## 6 0 0 0 0 1
## 18 0 0 0 0 0
## 21 0 0 0 0 0
## 26 0 0 0 0 0
## 31 0 0 0 0 1
## 90 0 0 0 0 0
## 98 0 1 0 0 0
## 102 1 0 0 0 0
## 169 0 0 0 0 0
## 187 0 0 0 0 0
## 532 0 0 0 0 0
## 677 0 0 0 0 0
## 743 0 0 0 0 0
## 750 0 0 1 0 0
## 792 0 0 0 0 0
## 1832 1 0 0 0 0
## T.docetaxel T.doxorubicin T.earli T.effect T.group T.iii T.metastat
## 6 0 0 0 0 1 0 0
## 18 0 0 0 0 0 0 0
## 21 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 0 1 0
## 90 0 1 0 0 0 0 1
## 98 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0
## 532 0 0 0 1 0 0 1
## 677 0 0 0 1 0 1 0
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 1 1
## 792 0 0 0 0 0 0 1
## 1832 0 0 0 0 0 0 0
## T.patient T.phase T.plus T.postmenopaus T.random T.randomis T.respons
## 6 1 0 0 0 0 0 0
## 18 0 0 0 0 0 0 0
## 21 0 0 1 0 0 0 1
## 26 1 0 0 0 0 0 0
## 31 1 1 0 0 1 0 0
## 90 0 0 0 0 1 0 0
## 98 0 0 0 0 0 0 1
## 102 0 0 0 0 0 0 0
## 169 0 1 0 0 0 0 0
## 187 1 1 0 0 0 1 0
## 532 1 0 0 0 0 0 0
## 677 1 1 0 0 0 0 0
## 743 1 1 0 0 0 0 1
## 750 0 1 0 0 0 0 0
## 792 0 1 1 0 0 0 0
## 1832 0 0 0 0 0 0 0
## T.result T.studi T.tamoxifen T.therapi T.treatment T.trial T.versus
## 6 0 1 0 0 0 1 0
## 18 0 0 0 0 0 0 0
## 21 0 0 1 1 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 0 1 1
## 90 0 0 0 0 1 1 0
## 98 0 0 0 1 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 1 0 0 0 0 0
## 187 0 0 0 0 0 1 0
## 532 0 0 0 0 0 0 0
## 677 0 0 0 0 0 1 0
## 743 1 1 0 0 0 0 0
## 750 1 1 0 0 0 1 0
## 792 0 0 0 0 1 1 0
## 1832 0 0 0 0 0 0 0
## T.women T.has.http T.num.chars T.num.words T.num.words.unq A.X0001
## 6 0 0 281 25 23 2
## 18 0 0 59 6 6 0
## 21 1 0 164 15 15 0
## 26 0 0 118 10 10 0
## 31 0 0 241 22 19 0
## 90 0 0 153 14 14 0
## 98 0 0 124 12 12 0
## 102 0 0 38 4 4 0
## 169 0 0 45 5 5 0
## 187 0 0 165 15 15 0
## 532 0 0 101 11 11 0
## 677 0 0 137 14 14 0
## 743 0 0 251 23 23 0
## 750 0 0 202 19 19 0
## 792 0 0 169 17 17 0
## 1832 0 0 103 9 9 0
## A.X001 A.X005 A.X100 A.X500 A.X5fluorouracil A.accord A.achiev
## 6 0 0 0 0 0 0 0
## 18 0 0 0 0 0 0 0
## 21 0 0 0 0 0 0 0
## 26 0 0 0 2 1 0 0
## 31 0 0 1 0 1 0 1
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 2
## 102 0 0 0 0 0 0 0
## 169 0 0 2 2 0 0 0
## 187 0 0 1 0 0 0 0
## 532 0 0 0 0 0 1 0
## 677 0 0 0 0 0 0 0
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 0 0
## 792 1 0 0 0 0 0 1
## 1832 0 0 0 0 0 0 0
## A.activ A.addit A.adjuv A.administ A.administr A.advanc A.advers
## 6 0 0 0 0 0 0 0
## 18 0 0 6 1 0 0 0
## 21 0 0 6 0 0 0 1
## 26 0 0 1 1 1 0 0
## 31 0 0 2 0 1 0 1
## 90 0 0 0 0 0 0 0
## 98 0 0 3 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 1 0 0 0 0 0 1
## 532 0 0 0 1 0 1 0
## 677 0 0 0 0 2 0 0
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 0 1
## 792 1 0 0 0 0 0 0
## 1832 0 0 5 0 0 0 0
## A.age A.agent A.aim A.alon A.also A.although A.among A.analys
## 6 1 0 0 0 0 0 0 3
## 18 1 0 0 2 0 0 0 0
## 21 0 0 0 0 0 0 0 1
## 26 0 0 0 0 0 0 0 0
## 31 0 0 0 0 1 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0 0
## 677 0 1 0 0 0 0 0 1
## 743 1 0 0 0 2 0 0 1
## 750 1 0 0 0 0 0 0 0
## 792 0 0 0 0 0 0 0 0
## 1832 1 0 0 0 0 0 0 0
## A.analysi A.analyz A.andor A.anthracyclin A.appear A.arm A.aromatas
## 6 0 0 0 0 0 0 0
## 18 0 0 0 0 0 0 0
## 21 0 0 0 0 0 0 0
## 26 2 0 0 0 0 0 0
## 31 0 0 0 0 0 2 0
## 90 0 0 0 0 0 0 0
## 98 1 0 0 0 0 1 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 2 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 0 0 0 0 0
## 743 0 0 0 0 0 0 0
## 750 1 0 0 0 0 14 0
## 792 1 0 0 0 0 0 0
## 1832 1 0 0 0 0 0 0
## A.assess A.assign A.associ A.avail A.axillari A.background A.base
## 6 0 0 1 0 0 0 0
## 18 0 0 1 0 1 0 0
## 21 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 0 1 0 0 1 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 1 0 0 0 0 1 0
## 532 0 0 0 0 0 0 0
## 677 0 0 2 0 0 1 0
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 1 0
## 792 2 0 0 1 0 1 0
## 1832 0 1 2 0 0 0 1
## A.baselin A.benefit A.better A.bone A.breast A.can A.cancer
## 6 0 1 0 0 1 0 1
## 18 0 0 0 0 3 0 1
## 21 0 1 0 0 1 0 0
## 26 0 0 0 0 1 0 1
## 31 0 1 0 0 2 0 2
## 90 0 0 0 0 0 0 0
## 98 0 1 0 0 2 0 2
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 12 0 4
## 532 0 0 0 0 1 1 1
## 677 0 0 0 0 1 0 3
## 743 0 0 0 0 2 0 2
## 750 0 0 0 0 2 0 1
## 792 0 3 0 0 1 1 1
## 1832 0 2 0 0 3 0 3
## A.carcinoma A.case A.caus A.cell A.chang A.characterist
## 6 0 0 0 0 0 0
## 18 2 1 0 0 0 0
## 21 0 0 0 0 0 0
## 26 0 0 0 0 0 0
## 31 0 0 0 0 0 0
## 90 0 0 0 0 0 0
## 98 0 0 0 0 0 0
## 102 0 0 0 0 0 0
## 169 0 0 0 0 1 0
## 187 0 0 0 0 2 0
## 532 0 0 0 0 0 0
## 677 0 0 0 0 0 0
## 743 0 0 0 1 0 0
## 750 0 0 0 0 0 0
## 792 0 0 0 0 0 0
## 1832 0 1 0 0 0 0
## A.chemotherapi A.clinic A.cmf A.combin A.common A.compar A.comparison
## 6 0 0 2 0 0 0 0
## 18 2 0 6 0 0 0 0
## 21 0 0 0 0 0 0 0
## 26 2 0 0 0 0 0 0
## 31 0 1 14 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 1 4 0 0 0 2 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 1 0 0
## 532 1 5 0 0 0 0 0
## 677 0 1 0 0 1 1 0
## 743 2 0 0 0 0 0 0
## 750 0 0 0 0 2 0 0
## 792 1 2 0 4 0 0 0
## 1832 3 0 3 0 0 2 1
## A.complet A.conclus A.conduct A.confid A.confirm A.consid A.consist
## 6 0 1 0 0 1 0 0
## 18 0 0 0 0 0 0 0
## 21 0 2 0 1 0 0 1
## 26 1 0 0 0 0 1 0
## 31 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 1 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 2 0 0 0 0 0 0
## 187 0 1 0 0 0 0 0
## 532 2 0 0 0 0 0 0
## 677 1 1 0 0 0 0 0
## 743 1 1 0 0 0 0 1
## 750 0 1 0 0 0 0 0
## 792 1 1 0 1 0 0 0
## 1832 0 1 1 0 0 0 0
## A.continu A.control A.correl A.cours A.cycl A.cyclophosphamid A.daili
## 6 0 0 0 1 0 1 0
## 18 0 0 0 0 0 1 0
## 21 0 1 0 0 0 0 0
## 26 0 1 0 1 1 1 0
## 31 0 0 0 0 3 1 0
## 90 0 0 0 0 0 0 0
## 98 1 0 0 0 3 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 1 0 0 0 0 0
## 532 1 0 0 0 0 0 0
## 677 0 0 0 0 0 0 1
## 743 0 0 3 0 1 0 0
## 750 0 0 0 0 0 0 0
## 792 0 0 0 0 1 0 1
## 1832 0 0 0 0 0 1 0
## A.data A.day A.death A.decreas A.defin A.demonstr A.design A.detect
## 6 0 0 0 0 0 0 0 0
## 18 0 0 0 0 0 0 0 1
## 21 1 0 0 0 0 0 0 0
## 26 0 7 0 0 0 0 0 0
## 31 0 2 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 0
## 187 0 1 0 0 0 0 0 0
## 532 1 0 0 0 0 0 0 0
## 677 0 1 0 0 0 0 0 0
## 743 1 3 0 0 0 0 1 0
## 750 2 4 0 0 0 1 0 0
## 792 0 3 0 0 0 0 0 0
## 1832 1 0 0 0 0 0 0 0
## A.determin A.develop A.dfs A.differ A.diseas A.diseasefre A.distant
## 6 1 0 3 1 0 1 2
## 18 0 0 0 1 0 0 0
## 21 0 0 5 0 0 2 0
## 26 0 0 0 2 0 0 0
## 31 0 0 1 2 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 0 4 1 0 1 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 2 0 0 0
## 187 0 0 0 1 0 0 0
## 532 2 1 0 0 0 0 0
## 677 0 0 0 0 1 0 0
## 743 0 0 0 1 0 0 0
## 750 0 0 0 1 1 0 0
## 792 0 0 0 1 0 0 0
## 1832 1 0 0 1 1 0 0
## A.docetaxel A.dose A.doubleblind A.doxorubicin A.drug A.due A.durat
## 6 0 0 0 0 0 0 0
## 18 0 0 0 0 0 1 0
## 21 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 1 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 2 0 0 0 0 0
## 187 0 0 1 0 1 0 0
## 532 0 1 0 0 0 0 0
## 677 0 5 0 0 0 1 0
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 0 0
## 792 0 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 0
## A.earli A.effect A.efficaci A.eight A.either A.elig A.end A.endocrin
## 6 0 1 0 0 0 0 0 0
## 18 0 0 0 0 0 0 0 0
## 21 0 0 0 0 0 0 0 3
## 26 0 0 0 0 1 0 1 0
## 31 1 1 0 0 1 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 1 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 0
## 187 3 1 2 0 0 1 0 0
## 532 0 4 0 0 0 0 0 0
## 677 0 6 0 0 1 0 0 0
## 743 0 1 0 0 0 0 0 0
## 750 0 0 0 0 0 0 0 0
## 792 0 0 0 0 0 0 0 0
## 1832 2 5 0 2 0 0 0 0
## A.endpoint A.enrol A.enter A.epirubicin A.estim A.estrogen A.evalu
## 6 0 0 0 0 1 0 0
## 18 0 0 1 0 0 0 1
## 21 0 0 0 0 2 2 2
## 26 0 0 1 1 0 0 1
## 31 0 1 0 1 0 0 1
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 2
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 2 0 0 0 0 0 0
## 532 0 0 0 0 0 0 1
## 677 2 1 0 0 0 0 0
## 743 0 0 1 0 0 0 2
## 750 0 0 0 0 0 0 1
## 792 0 1 0 0 0 0 0
## 1832 0 0 0 0 0 0 2
## A.event A.everi A.evid A.examin A.experienc A.express A.factor
## 6 0 0 0 1 0 0 0
## 18 0 0 1 0 0 0 0
## 21 0 0 0 0 0 0 1
## 26 0 0 0 0 1 0 0
## 31 1 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 0 0 0 0 0
## 743 0 1 0 0 0 1 0
## 750 2 2 0 0 0 0 1
## 792 0 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 0
## A.failur A.find A.first A.firstlin A.five A.fluorouracil A.follow
## 6 3 0 0 0 0 1 0
## 18 0 1 0 0 0 1 3
## 21 0 0 0 0 0 0 0
## 26 0 0 2 0 0 0 0
## 31 0 0 0 0 0 0 3
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 2
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 1
## 532 0 0 0 0 0 0 0
## 677 0 0 1 0 1 0 0
## 743 0 0 0 0 0 0 1
## 750 0 1 0 1 0 0 0
## 792 0 0 0 1 0 0 2
## 1832 0 0 0 0 0 1 0
## A.followup A.found A.four A.frequent A.function. A.general A.given
## 6 1 0 0 0 0 0 0
## 18 1 0 0 0 0 0 1
## 21 0 0 0 0 0 0 1
## 26 0 0 0 0 0 0 0
## 31 1 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 3 0 2 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 1 0 0
## 187 0 0 0 0 0 0 1
## 532 0 0 0 2 0 0 0
## 677 0 0 0 0 3 0 0
## 743 0 2 0 1 0 0 0
## 750 0 0 0 1 0 0 0
## 792 0 0 0 0 1 0 0
## 1832 2 0 0 0 0 0 0
## A.grade A.greater A.group A.growth A.hazard A.hematolog A.her2 A.high
## 6 0 0 1 0 1 0 0 0
## 18 0 0 0 0 0 0 0 0
## 21 0 1 0 0 2 0 0 0
## 26 0 0 0 0 0 0 0 0
## 31 1 0 0 0 2 0 0 1
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 1
## 187 0 0 2 0 0 0 0 0
## 532 0 0 0 1 0 0 0 3
## 677 1 0 0 0 0 0 0 1
## 743 1 0 0 0 0 0 0 0
## 750 0 0 0 1 0 0 0 0
## 792 0 0 0 0 0 0 0 1
## 1832 0 0 1 0 0 0 0 0
## A.higher A.histolog A.hormon A.howev A.human A.hundr A.identifi A.iii
## 6 0 0 0 2 0 0 1 0
## 18 0 0 0 0 0 0 0 0
## 21 0 0 0 0 0 0 0 0
## 26 0 0 0 1 0 0 0 0
## 31 2 1 0 0 0 0 1 1
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0 0
## 532 0 0 0 0 2 0 0 0
## 677 0 0 0 1 1 0 0 1
## 743 0 0 0 0 0 0 0 1
## 750 0 0 0 0 1 0 0 2
## 792 0 0 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 0 0
## A.import A.improv A.incid A.includ A.increas A.independ A.indic
## 6 1 0 0 0 0 2 1
## 18 0 0 0 0 1 0 0
## 21 0 0 0 0 0 0 0
## 26 0 0 0 2 0 0 0
## 31 1 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 1 0 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 0 0 2 0 0
## 743 0 0 0 1 0 0 0
## 750 0 0 0 1 0 0 0
## 792 0 0 0 1 0 0 0
## 1832 0 1 0 1 1 0 0
## A.inform A.infus A.inhibitor A.initi A.interv A.intraven A.investig
## 6 1 0 0 0 0 0 0
## 18 0 0 0 0 0 0 0
## 21 0 0 0 0 1 0 0
## 26 0 0 0 0 0 1 0
## 31 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 1 0 0 0 1 0
## 187 0 0 0 0 0 0 0
## 532 0 8 0 1 1 0 0
## 677 0 0 0 1 0 0 0
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 0 0
## 792 0 0 0 0 1 1 0
## 1832 0 0 0 0 0 0 0
## A.involv A.larg A.least A.less A.level A.life A.limit A.local
## 6 0 0 0 0 0 0 0 0
## 18 0 0 0 0 0 0 0 0
## 21 0 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0 0
## 31 0 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 2 0 0 0
## 187 0 0 0 0 0 0 0 0
## 532 0 0 0 0 6 0 0 0
## 677 0 0 0 0 0 0 0 0
## 743 1 0 0 0 0 0 0 0
## 750 0 0 0 0 0 0 0 1
## 792 0 0 0 0 0 2 0 0
## 1832 0 0 0 0 0 2 0 0
## A.longer A.low A.lower A.lymph A.main A.major A.marker A.mastectomi
## 6 0 0 0 1 0 0 0 2
## 18 0 0 0 0 0 0 0 1
## 21 0 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0 0
## 31 0 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 1 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0 0
## 677 0 1 0 0 0 0 0 0
## 743 0 0 1 0 0 0 3 0
## 750 0 0 0 0 0 0 0 0
## 792 0 0 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 0 0
## A.may A.mbc A.mean A.measur A.median A.metastas A.metastat A.method
## 6 0 0 0 0 1 0 0 1
## 18 1 0 0 0 1 0 0 0
## 21 1 0 0 0 0 0 0 2
## 26 0 0 0 0 0 0 0 0
## 31 0 0 0 0 1 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 1 0 0 1
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 0
## 187 0 0 1 1 0 0 0 1
## 532 0 0 0 2 0 0 0 0
## 677 1 0 0 2 0 0 0 1
## 743 1 0 0 0 1 0 0 1
## 750 0 0 0 0 0 0 1 1
## 792 0 1 0 0 3 0 1 1
## 1832 0 0 0 0 0 0 0 3
## A.methotrex A.mgm2 A.model A.month A.multicent A.multivari A.nausea
## 6 1 0 0 0 0 1 0
## 18 1 0 0 0 0 0 0
## 21 0 0 2 0 0 2 0
## 26 0 3 0 0 0 0 2
## 31 1 2 1 1 0 0 0
## 90 0 0 0 0 0 0 0
## 98 1 0 0 1 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 4 0 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 0 0 0 0 1
## 743 0 2 0 0 0 0 0
## 750 0 0 0 0 0 0 0
## 792 0 3 0 8 0 0 0
## 1832 1 0 0 2 0 0 0
## A.need A.negat A.neoadjuv A.neutropenia A.new A.node A.nodeposit
## 6 0 0 0 0 0 3 1
## 18 0 0 0 0 0 1 0
## 21 0 1 0 0 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 1 0 0 1
## 90 0 0 0 0 0 0 0
## 98 0 0 6 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 0 0 0 0 0
## 743 0 0 1 2 0 0 0
## 750 0 0 0 0 0 0 0
## 792 0 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 2
## A.number A.object A.observ A.obtain A.occur A.one A.oper A.oral
## 6 2 0 0 0 0 0 0 0
## 18 0 0 1 0 0 0 0 0
## 21 0 0 1 0 0 0 0 0
## 26 0 0 2 0 0 0 0 0
## 31 0 0 1 0 0 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 1 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 1 0 0 0 0 0
## 187 0 0 0 0 0 0 0 1
## 532 0 0 2 2 1 0 0 0
## 677 0 0 0 0 0 1 0 3
## 743 0 0 1 0 0 1 0 0
## 750 0 0 0 0 1 0 0 0
## 792 0 1 0 0 0 0 0 4
## 1832 0 0 0 0 0 0 0 0
## A.outcom A.overal A.paclitaxel A.partial A.particip A.patholog
## 6 0 1 0 0 0 1
## 18 0 0 0 0 0 0
## 21 0 3 0 0 1 0
## 26 0 0 0 0 0 0
## 31 1 0 0 0 0 0
## 90 0 0 0 0 0 0
## 98 0 1 0 0 0 0
## 102 0 0 0 0 0 0
## 169 0 0 0 0 0 0
## 187 0 0 0 0 0 0
## 532 0 0 0 0 0 0
## 677 0 0 0 0 0 0
## 743 0 1 1 0 0 2
## 750 0 0 3 0 0 0
## 792 0 2 0 0 0 0
## 1832 0 3 0 0 0 0
## A.patient A.per A.perform A.period A.phase A.placebo A.plus A.point
## 6 5 0 0 0 0 0 0 0
## 18 2 0 0 1 0 0 0 0
## 21 6 0 0 0 0 0 0 0
## 26 6 0 1 1 0 0 0 0
## 31 2 0 0 0 1 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 4 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 6 0 0 0 0 0 0 0
## 187 7 0 0 0 1 3 0 0
## 532 1 0 0 0 2 0 0 0
## 677 5 0 0 0 1 0 0 0
## 743 5 0 0 0 0 0 0 0
## 750 4 0 0 0 2 0 2 0
## 792 5 0 0 0 0 0 0 0
## 1832 6 0 1 0 0 0 0 0
## A.popul A.posit A.possibl A.postmenopaus A.postop A.potenti A.predict
## 6 0 3 0 0 0 0 0
## 18 0 1 0 0 1 0 0
## 21 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 1 1 0 1 0
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 0 0
## 792 0 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 0
## A.premenopaus A.present A.pretreat A.prevent A.previous A.primari
## 6 2 1 0 0 0 0
## 18 0 1 0 0 0 0
## 21 0 0 0 0 0 0
## 26 0 0 0 0 2 0
## 31 0 0 0 0 0 0
## 90 0 0 0 0 0 0
## 98 0 1 0 0 0 2
## 102 0 0 0 0 0 0
## 169 0 0 0 0 0 0
## 187 0 0 0 0 0 1
## 532 0 0 1 0 0 0
## 677 0 0 0 0 0 1
## 743 1 0 1 0 0 0
## 750 0 0 0 0 1 0
## 792 0 0 0 0 0 0
## 1832 0 0 0 0 0 0
## A.prior A.profil A.progesteron A.prognost A.progress A.progressionfre
## 6 0 0 0 2 0 0
## 18 0 0 0 0 0 0
## 21 0 0 0 1 0 0
## 26 0 0 0 0 0 0
## 31 0 0 0 0 0 0
## 90 0 0 0 0 0 0
## 98 0 0 0 0 0 0
## 102 0 0 0 0 0 0
## 169 0 0 0 0 0 0
## 187 0 0 0 0 0 0
## 532 0 0 0 0 0 0
## 677 2 0 0 0 0 0
## 743 0 0 0 0 0 0
## 750 0 1 0 0 1 0
## 792 0 1 0 0 0 1
## 1832 0 0 0 0 0 0
## A.prolong A.proport A.prospect A.provid A.purpos A.qualiti
## 6 0 0 0 0 1 0
## 18 0 0 0 0 0 0
## 21 0 1 0 0 1 0
## 26 0 0 0 0 0 0
## 31 0 0 0 0 0 0
## 90 0 0 0 0 0 0
## 98 0 0 1 0 0 0
## 102 0 0 0 0 0 0
## 169 0 0 0 0 0 0
## 187 0 0 0 0 2 0
## 532 0 0 0 0 0 0
## 677 0 0 0 0 0 0
## 743 0 0 0 0 0 0
## 750 0 0 0 0 0 0
## 792 0 0 0 0 1 2
## 1832 0 0 0 2 2 2
## A.radiotherapi A.random A.randomis A.rang A.rate A.ratio A.receiv
## 6 2 1 0 0 1 1 0
## 18 0 1 0 0 0 0 1
## 21 0 1 0 0 0 1 0
## 26 0 0 1 0 0 0 2
## 31 0 1 0 0 1 2 1
## 90 0 0 0 0 0 0 0
## 98 0 0 2 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 4
## 187 4 0 3 0 0 0 0
## 532 0 0 0 0 0 0 1
## 677 0 0 0 0 0 0 0
## 743 0 0 0 1 1 0 0
## 750 0 0 2 0 0 0 0
## 792 0 0 0 3 1 0 3
## 1832 0 1 0 0 0 0 1
## A.receptor A.receptorposit A.recurr A.reduc A.reduct A.regard
## 6 0 0 0 0 0 0
## 18 0 0 0 0 0 0
## 21 0 2 0 0 0 0
## 26 0 0 0 0 0 1
## 31 0 0 0 0 0 0
## 90 0 0 0 0 0 0
## 98 0 0 0 0 0 0
## 102 0 0 0 0 0 0
## 169 0 0 0 0 0 0
## 187 0 0 0 0 1 0
## 532 0 0 0 0 0 0
## 677 0 0 0 0 0 0
## 743 0 0 0 0 0 0
## 750 0 0 1 0 0 0
## 792 0 0 0 0 0 1
## 1832 0 0 0 0 0 0
## A.regimen A.regress A.relaps A.relat A.remain A.report A.requir
## 6 0 0 0 0 1 0 0
## 18 0 0 0 0 0 0 0
## 21 0 1 0 0 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 1 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 0 1 0 0 0
## 743 1 0 0 0 0 0 0
## 750 1 0 0 0 0 3 0
## 792 0 0 0 0 0 0 0
## 1832 0 0 2 0 0 0 0
## A.respect A.respond A.respons A.result A.risk A.safeti A.sampl
## 6 0 0 0 2 0 0 0
## 18 1 0 0 0 1 0 0
## 21 2 0 3 1 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 1 2 4 1 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 1 0 0 0 0
## 187 0 0 0 1 0 0 0
## 532 2 0 0 1 0 0 1
## 677 0 0 0 1 0 1 0
## 743 0 0 4 2 0 0 0
## 750 0 0 0 1 0 4 0
## 792 1 0 2 1 0 1 0
## 1832 0 0 0 1 0 0 0
## A.schedul A.score A.second A.secondari A.seen A.select A.sequenti
## 6 0 0 0 0 0 0 0
## 18 0 0 3 0 0 0 0
## 21 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 1 0 1
## 90 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 1 0 0
## 187 0 0 0 1 0 0 0
## 532 0 0 0 0 0 0 0
## 677 0 0 1 0 0 0 1
## 743 0 0 0 0 0 0 0
## 750 0 0 0 0 0 0 0
## 792 2 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 0
## A.serum A.set A.seven A.sever A.show A.shown A.side A.signific
## 6 0 0 0 0 0 0 0 3
## 18 0 0 0 0 0 0 0 0
## 21 0 0 0 0 1 0 0 0
## 26 0 0 0 0 0 0 0 2
## 31 0 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0 1
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 1
## 187 0 0 0 0 1 0 0 1
## 532 6 0 0 0 0 1 2 0
## 677 0 0 0 0 1 0 1 1
## 743 0 0 0 0 2 0 0 1
## 750 0 0 0 0 0 0 0 0
## 792 0 0 0 0 2 0 0 1
## 1832 0 0 0 0 0 1 1 0
## A.similar A.singl A.site A.six A.size A.stabl A.stage A.standard
## 6 0 0 0 1 1 0 0 0
## 18 0 0 0 0 0 0 0 0
## 21 0 0 0 0 0 0 0 0
## 26 0 0 0 0 0 0 0 0
## 31 1 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 0 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0 0
## 187 0 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0 0
## 677 0 0 0 0 0 0 0 0
## 743 0 1 0 1 0 0 2 0
## 750 1 0 0 0 0 0 0 0
## 792 0 0 0 0 0 0 0 0
## 1832 0 0 0 0 0 0 0 0
## A.start A.statist A.status A.studi A.subgroup A.suggest A.superior
## 6 0 1 0 0 2 0 0
## 18 0 0 0 1 0 1 0
## 21 0 0 2 2 1 0 0
## 26 0 0 0 1 0 0 0
## 31 0 0 0 1 1 0 2
## 90 0 0 0 0 0 0 0
## 98 0 1 0 0 0 0 2
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 2 0 0 0
## 532 1 0 0 2 0 1 0
## 677 0 1 0 2 0 0 0
## 743 0 0 0 1 0 1 0
## 750 0 0 0 1 0 0 0
## 792 0 1 0 1 0 1 0
## 1832 0 0 0 0 0 0 0
## A.support A.surgeri A.surviv A.system A.tamoxifen A.term A.test
## 6 1 1 2 0 0 0 0
## 18 0 4 0 0 0 0 0
## 21 0 0 4 0 2 0 1
## 26 0 0 0 0 0 0 0
## 31 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0
## 98 1 2 3 0 1 0 0
## 102 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 0
## 187 0 0 0 0 0 1 1
## 532 0 0 0 0 0 0 0
## 677 0 0 0 0 0 0 0
## 743 0 2 1 0 0 0 0
## 750 0 0 0 0 0 0 0
## 792 0 0 2 0 0 0 1
## 1832 0 0 6 1 0 3 0
## A.therapi A.three A.time A.tissu A.toler A.total A.toxic A.treat
## 6 0 1 0 0 0 0 0 2
## 18 0 0 0 0 0 2 0 0
## 21 3 0 0 0 0 0 0 0
## 26 0 0 0 0 1 0 0 0
## 31 0 0 0 0 0 0 0 0
## 90 0 0 0 0 0 0 0 0
## 98 2 0 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 0 0 0 1 0
## 187 0 1 1 3 0 0 0 0
## 532 0 0 1 0 2 0 6 0
## 677 0 0 0 0 1 0 1 0
## 743 2 0 0 1 1 0 1 0
## 750 0 0 0 0 0 0 1 0
## 792 0 0 0 0 0 0 2 0
## 1832 1 0 5 0 0 0 6 1
## A.treatment A.trend A.trial A.tumor A.tumour A.two A.type A.use
## 6 0 0 0 1 0 0 1 1
## 18 1 0 0 2 0 1 0 0
## 21 3 0 1 5 0 0 0 1
## 26 2 0 0 0 0 3 0 1
## 31 0 0 0 2 0 1 0 0
## 90 0 0 0 0 0 0 0 0
## 98 3 0 1 0 0 0 0 1
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 1 0 0 0 0
## 187 1 0 1 0 0 0 0 0
## 532 0 0 2 2 0 1 0 0
## 677 1 0 1 0 0 1 0 2
## 743 2 0 1 0 2 1 0 0
## 750 0 0 1 0 0 0 0 0
## 792 1 0 8 0 0 0 0 0
## 1832 5 0 2 0 0 0 0 1
## A.valu A.versus A.vomit A.week A.well A.wherea A.whether A.within
## 6 1 1 0 0 0 0 2 2
## 18 0 0 0 0 0 0 0 0
## 21 1 0 0 0 0 0 0 0
## 26 0 0 2 0 0 0 0 0
## 31 0 0 0 0 0 0 1 0
## 90 0 0 0 0 0 0 0 0
## 98 0 6 0 0 0 0 0 0
## 102 0 0 0 0 0 0 0 0
## 169 0 0 0 1 0 0 0 0
## 187 0 0 0 0 0 0 0 0
## 532 0 0 0 0 0 0 0 1
## 677 0 0 0 1 1 0 1 0
## 743 0 0 0 0 0 0 0 0
## 750 0 2 0 0 0 0 0 0
## 792 0 0 0 1 0 0 0 0
## 1832 1 1 0 0 0 0 1 2
## A.without A.women A.year A.has.http A.num.chars A.num.words
## 6 2 1 2 0 1738 146
## 18 0 1 3 0 1403 113
## 21 0 0 0 0 1728 146
## 26 1 0 0 0 1329 106
## 31 0 0 0 0 1501 139
## 90 0 0 0 0 0 0
## 98 0 1 2 0 1346 122
## 102 0 0 0 0 0 0
## 169 0 0 0 0 628 60
## 187 0 0 1 0 1666 156
## 532 2 0 0 0 2223 193
## 677 0 0 0 0 1751 163
## 743 0 0 1 0 1711 146
## 750 0 0 0 0 1316 131
## 792 0 1 0 0 1762 164
## 1832 2 2 6 0 2028 187
## A.num.words.unq trial.fctr.predict.Final.rf.prob
## 6 105 0.398
## 18 80 0.408
## 21 87 0.206
## 26 67 0.566
## 31 103 0.794
## 90 0 0.138
## 98 72 0.408
## 102 0 0.428
## 169 38 0.514
## 187 90 0.502
## 532 98 0.284
## 677 117 0.544
## 743 109 0.838
## 750 89 0.814
## 792 107 0.778
## 1832 106 0.472
## trial.fctr.predict.Final.rf trial.fctr.predict.Final.rf.accurate
## 6 N TRUE
## 18 Y FALSE
## 21 N FALSE
## 26 Y FALSE
## 31 Y TRUE
## 90 N FALSE
## 98 Y FALSE
## 102 Y FALSE
## 169 Y FALSE
## 187 Y FALSE
## 532 N TRUE
## 677 Y FALSE
## 743 Y FALSE
## 750 Y FALSE
## 792 Y FALSE
## 1832 Y FALSE
## .label
## 6 .6
## 18 .18
## 21 .21
## 26 .26
## 31 .31
## 90 .90
## 98 .98
## 102 .102
## 169 .169
## 187 .187
## 532 .532
## 677 .677
## 743 .743
## 750 .750
## 792 .792
## 1832 .1832
tmp_replay_lst <- replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.new.prediction")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
## 3.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: data.training.all.prediction
## 4.0000 5 0 1 1 1
## 4.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: model.final
## 5.0000 4 0 0 2 1
## 6.0000 6 0 0 1 2
print(ggplot.petrinet(tmp_replay_lst[["pn"]]) + coord_flip())
Null Hypothesis (\(\sf{H_{0}}\)): mpg is not impacted by am_fctr.
The variance by am_fctr appears to be independent. #{r q1, cache=FALSE} # print(t.test(subset(cars_df, am_fctr == "automatic")$mpg, # subset(cars_df, am_fctr == "manual")$mpg, # var.equal=FALSE)$conf) # We reject the null hypothesis i.e. we have evidence to conclude that am_fctr impacts mpg (95% confidence). Manual transmission is better for miles per gallon versus automatic transmission.
## chunk_label chunk_step_major chunk_step_minor elapsed
## 11 fit.models 5 2 426.511
## 13 fit.data.training.all 6 1 511.814
## 10 fit.models 5 1 89.153
## 7 select_features 4 0 26.089
## 12 fit.data.training.all 6 0 440.471
## 9 fit.models 5 0 40.403
## 14 predict.data.new 7 0 517.551
## 6 extract.features 3 0 6.239
## 8 remove_correlated_features 4 1 27.537
## 2 cleanse_data 2 0 1.422
## 4 manage_missing_data 2 2 2.694
## 5 encodeORretype.data 2 3 2.973
## 3 inspectORexplore.data 2 1 1.504
## 1 import_data 1 0 0.002
## elapsed_diff
## 11 337.358
## 13 71.343
## 10 48.750
## 7 19.850
## 12 13.960
## 9 12.866
## 14 5.737
## 6 3.266
## 8 1.448
## 2 1.420
## 4 1.190
## 5 0.279
## 3 0.082
## 1 0.000## [1] "Total Elapsed Time: 517.551 secs"
## R version 3.1.3 (2015-03-09)
## Platform: x86_64-apple-darwin13.4.0 (64-bit)
## Running under: OS X 10.10.3 (Yosemite)
##
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
##
## attached base packages:
## [1] tcltk grid stats graphics grDevices utils datasets
## [8] methods base
##
## other attached packages:
## [1] randomForest_4.6-10 rpart.plot_1.5.2 rpart_4.1-9
## [4] ROCR_1.0-7 gplots_2.16.0 caret_6.0-41
## [7] lattice_0.20-31 tm_0.6 NLP_0.1-6
## [10] sqldf_0.4-10 RSQLite_1.0.0 DBI_0.3.1
## [13] gsubfn_0.6-6 proto_0.3-10 reshape2_1.4.1
## [16] plyr_1.8.1 caTools_1.17.1 doBy_4.5-13
## [19] survival_2.38-1 ggplot2_1.0.1
##
## loaded via a namespace (and not attached):
## [1] bitops_1.0-6 BradleyTerry2_1.0-6 brglm_0.5-9
## [4] car_2.0-25 chron_2.3-45 class_7.3-12
## [7] codetools_0.2-11 colorspace_1.2-6 compiler_3.1.3
## [10] digest_0.6.8 e1071_1.6-4 evaluate_0.5.5
## [13] foreach_1.4.2 formatR_1.1 gdata_2.13.3
## [16] gtable_0.1.2 gtools_3.4.1 htmltools_0.2.6
## [19] iterators_1.0.7 KernSmooth_2.23-14 knitr_1.9
## [22] labeling_0.3 lme4_1.1-7 MASS_7.3-40
## [25] Matrix_1.2-0 mgcv_1.8-6 minqa_1.2.4
## [28] munsell_0.4.2 nlme_3.1-120 nloptr_1.0.4
## [31] nnet_7.3-9 parallel_3.1.3 pbkrtest_0.4-2
## [34] quantreg_5.11 RColorBrewer_1.1-2 Rcpp_0.11.5
## [37] rmarkdown_0.5.1 scales_0.2.4 slam_0.1-32
## [40] SparseM_1.6 splines_3.1.3 stringr_0.6.2
## [43] tools_3.1.3 yaml_2.1.13